AfterAccessICT access and use in Asiaand the Global South

V E R S I O N 3 . 0 — A P R I L 2 0 1 9

A report based on nationally representative surveys of households and individuals conducted by DIRSI, LIRNEasia and Research ICT Africa

AfterAccessICT access and use in Asia

and the Global South

AfterAccess

2

About AfterAccess | The AfterAccess surveys are conducted by pro-poor sister networks DIRSI, LIRNEasia and Research ICT Africa. The surveys are nationally representative and use methodology that is comparable across the countries. AfterAccess currently includes completed surveys in 23 countries across the Global South – seven in Asia, ten in Africa and six in Latin America – making it the most comprehensive database on mobile phone and Internet access and use in the Global South. For more information visit afteraccess.net or follow @AfterAccess.

About LIRNEasia | LIRNEasia is a pro-poor, pro-market think tank. Its mission is catalyzing policy change through research to improve people’s lives in the emerging Asia Pacific by facilitating their use of hard and soft infrastructures through the use of knowledge, information and technology. LIRNEasia has been active in the Asia Pacific since 2005, conducting both demand- and supply-side research as well as advocating for policy changes in the ICT sector on issues ranging from universal service policy to open data, gender, big data and more. For more information, visit lirneasia.net or follow @LIRNEasia.

Funding | The research presented in this report was conducted with financial support from the International Development Research Centre (IDRC) Canada, the Ford Foundation and the Swedish International Development Cooperation Agency (SIDA).

Disclaimer | The views expressed in this work are the views of the authors and do not necessarily represent those of the International Development Research Centre (IDRC) Canada or its Board of Governors, the Ford Foundation, the Swedish International Development Cooperation Agency (SIDA) or any of the consultants hired to conduct the fieldwork.

Suggested citation | LIRNEasia (2019). AfterAccess: ICT access and use in Asia and the Global South (Version 3.0). Colombo: LIRNEasia

©LIRNEasia, March 2019

Asia Report 3.0

3

Core Research Team | To date, the following people from the three research organizations have contributed to the AfterAccess project:

LIRNEasia | Helani Galpaya, Ayesha Zainudeen, Tharaka Amarasinghe, Laleema Senanayake, Firas Mohamed, Suthaharan Perampalam, Gayani Hurulle, Namali Premawardhana, Ranjula Senaratne-Perera, Radhika Gunewardena. Research ICT Africa | Alison Gillwald, Mothobi Onkokame, Mariama Deen-Swarray, Christoph Stork, Broc Rademan, Enrico Calandro, Chenai Chair, Anri van der Spuy. DIRSI | Roxana Barrantes, Aileen Aguero, Paulo Matos, Diego Aguilar, Gera Rios, Greta Zamora.Acknowledgements | The authors wish to acknowledge all those who provided invaluable input and engaged with LIRNEasia and the wider AfterAccess team on the design, implementation, and analysis of the survey.

Mention must be made of those who provided input on the design of earlier versions of the survey which were implemented in Myanmar: Jorge Garcia Hombrados, Srignesh Lokanathan, Nilusha Kapugama, Joshua Blumenstock, Per Helmerson and Saad Gulzar. The advice given by Rohan Samarajiva, who also spent time reviewing the early part of data analysis, is appreciated.

The assistance of the Pakistan Telecom Authority, in enabling the research to be conducted in Pakistan, is appreciated.

AfterAccess

4

Asia Report 3.0

5

List of abbreviations 6

About the study 7

About AfterAccess 8 AfterAccess base methodology 9 Sri lanka survey and sampling methodology 12 India survey and sampling methodology 13 Pakistan survey and sampling methodology 14 Myanmar survey and sampling methodology 15 Bangladesh survey and sampling methodology 16 Cambodia survey and sampling methodology 17 Nepal survey and sampling methodology 18

Sample size determination 19

Weighting of data 19

Note on reading this report 20

Connectivity 21 Mobile phone ownership 22 Type of handset 27 New adopters 31 Multiple sim ownership 33 Internet 35 App use 47

Social media 49

Public Wi-Fi 61

Mobile phone expenditure 67

Online harassment 71

E-commerce 77 Mobile money 78 Platforms 79 Using platforms for buying goods and services 79 Using platforms for selling goods and services 87

Cybersecurity 93

Table ofcontents

AfterAccess

6

List of abbreviationsAJK Azad, Jammu and KashmirARPU Average revenue per userEA Enumerator areaFATA Federally administered tribal areasGND Grama niladhari divisionGNI Gross national incomeGPS Global positioning systemICT Information and communication technologyPPP Purchasing power parityPPS Probability proportionate to sizePSU Primary sampling unitSEC Socio-economic classificationUSD United States dollars

Asia Report 3.0

7

About the study

AfterAccess

8

About AfterAccess

AfterAccess is a series of surveys on how individuals in 23 countries of the Global South access and use ICTs. The surveys reported here were conducted between April 2017 and January 2019 across Asia, Africa and Latin America, with the exception of the Myanmar survey which took place mid-2016. The research was conducted via 46,746 face-to-face household and individual interviews lasting 90 minutes each.

The objective of this global effort is to collect a range of household and individual data that can offer much greater insight on the demand-side barriers to digital equality, and in this way provide comprehensive bases of evidence to inform national and regional policy and regulation.

The surveys cover a wide range of topics related to the use of mobile phones, Internet, social media and other platforms. AfterAccess is uniquely positioned to disrupt the current narratives of homogeneity in mobile phone and Internet access and use, illustrate the multifaceted

challenges faced by the developing world and identify precise points of policy intervention.

The surveys are nationally-representative of the 15-65 age group in each country, with a confidence level of 95% and a margin of error of approximately ±3%. That is, the data can be extrapolated to the 15-65 population in each country to produce estimates which will be within ±3% of the actual levels. The methodology and sample sizes allow for disaggregated analysis of urban-rural populations, genders and socio-economic groups (SEC) at the national level. Sample sizes vary from 1,171 in Mozambique to 5,069 in India (Figure 1).

The research was conducted by LIRNEasia in Asia, DIRSI in Latin America and Research ICT Africa in Africa. Comparable methodology and a (predominantly) common questionnaire were used across regions and countries.

AfterAccess has been recognized by the EQUALS global partnership for its contribution to bringing rigorous data to the policy process, specifically in the area of gender equality in technology. AfterAccess was selected from over 350 nominees for the 2018 Research category award. The award is intended to recognize ‘outstanding projects and initiatives around the world that are helping women and girls become #EQUALSinTech.’

Asia Report 3.0

9

AfterAccess base methodology

The key objective of the methodology is to ensure national representation at the desired levels of precision. The main requirements for this are:

(1) a comprehensive national sample frame at the most granular level possible (census enumerator areas (EAs) or blocks in the best case1); and

(2) random selection at every level of sample selection (i.e.: EA, household, individual)

The AfterAccess base methodology is designed to meet these two requirements.

1 A census divides a country in to blocks or EAs which have a rough density of 200 households. This is generally considered a manageable number of households that can be listed within a day.2 Aged 15+ in Africa and Latin America

AfterAccess base methodology

(1) Separation of EA sample frame into urban and rural EAs(2) Sampling the required number of EAs from each stratum (urban and rural) using probability proportionate to size (PPS)(3) Mapping, listing and marking all households in the selected EA The lists serve as the sample frame for simple random selection of households. This was done with the assistance of key informants (e.g.: ward/ village leader, etc.) (4) Simple random selection of the required number of households (20-25) from each selected EA(5) Listing all household members or visitors aged 15-652 staying the night at the selected household(6) Simple random selection (using the Kish grid) of one household member for individual survey from household list compiled in Step (5)

COLOMBIA

KENYA

1,208

TANZANIA

1,200

RWANDA

1,211

MOZAMBIQUE

1,171

SOUTH AFRICA

1,815

LESOTHO

GHANA

1,200

NIGERIA

1,808

SENEGAL

1,233

UGANDA

1,864

PAKISTAN

2,002

INDIA

5,069CAMBODIA

2,123

BANGLADESH

2,020

SRI LANKA

2.017

NEPAL

2,008

MYANMAR

7,500

GUATEMALA

ECUADOR

1,517

ARGENTINA

1,500

PERU

1,544

1,539

PARAGUAY

1,5701,500

2,127 Figure 1. AfterAccess countries covered and sample sizes. Pakistan excludes AJK, FATA and Gilgit-Baltistan (~2% of population)

AfterAccess

10

Figure 2. Listing of households in Pakistan

Figure 5. Key informant interview in IndiaFigure 4. Listing of households in IndiaFigure 3. Key informant interview in Cambodia

Asia Report 3.0

11

In the AfterAccess base methodology, the random sampling of households and individuals (and other target groups that may be added on) is based on the census enumerator area sampling frames. In each country, the base methodology outlined above was adjusted depending on the availability and granularity of sample frames, and ground realities. The lowest administrative level sampling frames available to the public were grama niladhari divisions (GND) in Sri Lanka, villages and wards in India and Bangladesh, wards and village tracts in Myanmar, villages in Cambodia and wards in Nepal. No sampling frame was publicly available for Pakistan. Hence, the Pakistan Bureau of Statistics conducted the sampling of EAs from the national census sampling frame for us, using the AfterAccess base methodology and following our specifications.

Where the EA sampling frame was not available, the lowest-level administrative units publicly available for each country (ward, village, village-tract or GNDs as appropriate) were divided into smaller areas for listing and enumeration. These administrative units typically have a larger number of households than an EA. For instance, some wards (specifically in Mumbai) can have as many as 100,000 households, making listing households (Step 3) impossible if selected into the sample. Therefore, large administrative units were segmented on the field, according to pre-defined methodology, and

one or more smaller segments then randomly selected for listing and enumeration (Step 4 onwards). These segments were then treated as EAs and Steps 3 onwards were applied accordingly.

In India, additional steps were required to balance the twin priorities of capturing the diversity of the population and managing fieldwork costs in the vast nation. These are described in detail on page 13. A larger sample size was also implemented in India, in consideration of these factors.

Country-specific summaries are provided in the following sub-sections, but it is important to note that the core principle of random selection was incorporated at every stage of sample selection to ensure national representation. There was no purposive, convenience or quota selection of any kind.

The fieldwork was conducted in Asia by competitively-selected market research companies. The companies were mainly involved in the fieldwork set-up (including scripting, translating and pilotd-testing of the questionnaire and training of enumerators) and execution as well as dataset delivery. LIRNEasia monitored the companies, in most cases by participating in field training and monitoring fieldwork both on the ground as well as remotely. The sample sizes in each country allow for disaggregation of

data by urban-rural, gender and SEC at the national level. The data cannot be analyzed by province, state, district, sub district etc. in any of the countries except Myanmar, as the samples were not designed to do so. In Myanmar, state-level disaggregation is possible, as the sample was designed to do so.

For detailed notes on methodology-please visit: bit.ly/AAAmethod

Figure 7. Segmentation map, Bangladesh

Figure 6. Segmentation map, India

AfterAccess

12

Sri Lanka survey and sampling methodology

In Sri Lanka, 2,017 households and individuals were surveyed from 100 GNDs in all nine (9) provinces (Figure 8). The sampling methodology was designed to ensure representation of the target group (population aged 15-65) at a national level with a 95% confidence interval and a ±3.3% margin of error. That is, the data can be extrapolated to the 15-65 population of Sri Lanka with statistical confidence. The methodology and sample sizes also allow for disaggregated analysis of urban-rural populations and genders at the national level. The data cannot be analyzed by province, district or GND as the sample was not designed to do so.

The method was developed using GND-level data from the National Census of Population and Housing 2012. Sampling was conducted by LIRNEasia, using the AfterAccess base methodology with segmentation of wards as described earlier.

Fieldwork was conducted by Nielsen Sri Lanka in November 2018 – January 2019, with supervision of all field activities by LIRNEasia on the ground as well as remotely. The data was collected using mobile devices, and uploaded and reviewed on a daily basis, with live monitoring of GPS locations of survey teams.

Raw data collected was then weighted using 2018 mid-year national population estimates to correct for over- and under-sampling of certain population sub-groups. Figure 8. Sri Lanka sample locations based on

GPS coordinates recorded during fieldwork

Asia Report 3.0

13

India survey and sampling methodology

In India, 5,069 households and individuals were surveyed from 250 wards and villages in 19 states and 108 districts (Figure 9).

The sampling methodology was designed to ensure representation of the target group (population aged 15-65) at a national level with a 95% confidence interval and a ±2.8% margin of error. That is, the data can be extrapolated to the 15-65 population of India with statistical confidence. The methodology and sample sizes also allow for disaggregated analysis of urban-rural populations and genders at the national level. The data cannot be analyzed by state, district or sub-district as the sample was not designed to do so.

The method was developed using (urban) ward- and (rural) village-level data from the 2011 National Primary Census Abstract Data, compiled from the Census of India website. Sampling was conducted by LIRNEasia, using the AfterAccess base methodology with segmentation of wards and villages as described earlier.

Two steps were added to the AfterAccess base methodology

to manage the geographic spread of the sample in India while maintaining representativeness/randomness. Before wards and villages were selected, first, districts were randomly selected, and second, within the selected districts, sub-districts were randomly selected. Thereafter, wards and villages were randomly selected from the selected sub-districts. As stated earlier, randomness was maintained at every level of selection, thus ensuring representativeness.

Field set-up, execution and dataset delivery were conducted by Ipsos India in October-November 2017, with supervision of all field activities by LIRNEasia. The data was collected using mobile devices, and uploaded and reviewed on a daily basis, with live monitoring of GPS locations of survey teams.

Raw data collected was then weighted using 2017 national population estimates to correct for over- and under-sampling of certain population sub-groups.

Figure 9. India sample locations based on GPS coordinates recorded during fieldwork

AfterAccess

14

Pakistan survey and sampling methodology

In Pakistan, 2,002 households and individuals were surveyed from 100 census enumerator areas (Figure 10).

The sampling methodology was designed to ensure representation of 98% of the target group (population aged 15-65) at a national level with a 95% confidence interval and a ±3.3% margin of error. That is, the data can be extrapolated to the 15-65 population of Pakistan with statistical confidence. The methodology and sample sizes also allow for disaggregated analysis of urban-rural populations and genders in Pakistan at the national level. The data cannot be analyzed by province as the sample was not designed to do so.

Since the national census sampling frame is not publicly accessible, the Pakistan Bureau of Statistics (PBS) provided a sample of 100 EAs (selected according to the AfterAccess methodology) and facilitated access to the maps for the selected EAs. Sampling was based on the 2017 Census of Pakistan sampling frame. The AJK, FATA and Gilgit-Baltistan provinces – amounting to approximately 2% of the

Figure 10. Pakistan sample locations based on GPS coordinates recorded during fieldwork

population – were excluded from the sample frame due to practical and security considerations.

Field set-up, execution and dataset delivery were conducted by The Dynamics Research in October-December 2017, with supervision of all field activities by LIRNEasia. The data was collected using mobile devices, and uploaded and reviewed on a daily basis, with live monitoring of GPS locations of survey teams.

Raw data collected was then weighted using 2017 national population data to correct for over- and under-sampling of certain population sub-groups.

Asia Report 3.0

15

Myanmar survey and sampling methodology

The Myanmar survey of 2016 was conducted as a follow up to a baseline survey of ICT use in 2015, to track progress after liberalization of the country’s mobile sector in 2013. As such, the sampling process was somewhat different and sample size was much larger, to enable national representation and state-wise disaggregation of the data at the same time.

In Myanmar, 7,500 households and individuals were surveyed from 72 townships (Figure 11). The survey was conducted in June-August 2016.

The sampling methodology was designed to ensure representation of the target group (population aged 15-65 living in accessible areas – approximately 97% of households or 96.3% of the population) at a national level with a 95% confidence interval and a ±3% margin of error. That is, the data can be extrapolated to the 15-65 population of Myanmar with statistical confidence.

The township was set as a PSU. Thirty-two of the 330 townships (containing approximately 3% of the households

Figure 11. Myanmar sample locations based on GPS coordinates recorded during fieldwork

and population aged 15-65) were excluded from the sampling frame due to inaccessibility and security concerns.

Townships were stratified into geographic areas, population-based strata, as well as urban and rural. Seventy townships were selected with PPS systematic sampling. Within townships, two wards were selected in urban and four village tracts in rural areas. Within each, two segments (streets) were then selected using PPS systematic sampling. Following this, households were selected from a local administrator’s household list, using systematic random sampling. Within a selected household, the Kish grid was used to select a member within the 15-65 age group, for survey.

The results can be disaggregated into administrative regions and states, geographic regions, urban versus rural locations, as well as by gender and age groups.

Fieldwork was conducted by Third Eye Co. Ltd. with supervision of all field activities by LIRNEasia.

Raw data collected was then weighted using 2016 national population estimates to correct for over- and under-sampling of certain population sub-groups.

AfterAccess

16

Bangladesh survey and sampling methodology

In Bangladesh, 2,020 households and individuals were surveyed from 100 wards and villages in 40 Zillas (Figure 12).

The sampling methodology was designed to ensure representation of the target group (population aged 15-65) at a national level with a 95% confidence interval and a ±3.3% margin of error. That is, the data can be extrapolated to the 15-65 population of Bangladesh with statistical confidence. The methodology and sample sizes also allow for disaggregated analysis of urban-rural populations and genders at the national level. The data cannot be analyzed by Zilla, as the sample was not designed to do so.

The method was developed using (urban) ward- and (rural) village-level data from the 2011 National Census Data. Sampling was conducted by LIRNEasia using the AfterAccess base methodology, with segmentation of wards and villages as described earlier.

Fieldwork was conducted in October-November 2017 by Ipsos India with supervision of all field activities by LIRNEasia. The data was collected

using mobile devices, and uploaded and reviewed on a daily basis, with live monitoring of GPS locations of survey teams.

Raw data collected was then weighted using 2017 national population estimates to correct for over- and under-sampling of certain population sub-groups.

Figure 12. Bangladesh sample locations based on GPS coordinates recorded during fieldwork

Asia Report 3.0

17

Cambodia survey and sampling methodology

In Cambodia, 2,123 households and individuals were surveyed from 100 villages in 25 provinces (Figure 13).

The sampling methodology was designed to ensure representation of the target group (population aged 15-65) at a national level, with a 95% confidence interval and a ± 3.3% margin of error. That is, the data can be extrapolated to the 15-65 population of Cambodia with statistical confidence. The methodology and sample sizes also allow for disaggregated analysis of urban-rural populations and genders at the national level. The data cannot be analyzed by province as the sample was not designed to do so.

The method was developed using village-level data from the 2014 inter-censual survey. Sampling was conducted by LIRNEasia using the AfterAccess base methodology, with segmentation of villages as described earlier.

Fieldwork was conducted in September-October 2017 by Kantar TNS Cambodia with supervision of all

Figure 13. Cambodia sample locations based on GPS coordinates recorded during fieldwork

field activities by LIRNEasia. The data was collected using mobile devices and uploaded and reviewed on a daily basis with live monitoring of GPS locations of survey teams.

Raw data collected was then weighted using 2017 national population estimates to correct for over- and under-sampling of certain population sub-groups.

AfterAccess

18

Nepal survey and sampling methodology

In Nepal, 2,008 households and individuals were surveyed from 100 wards in 7 provinces (Figure 14).

The sampling methodology was designed to ensure representation of the target group (population aged 15-65) at a national level with a 95% confidence interval and a ±3.3% margin of error. That is, the data can be extrapolated to the 15-65 population of Nepal with statistical confidence. The methodology and sample sizes also allow for disaggregated analysis of urban-rural populations and genders at the national level. The data cannot be analyzed by province as the sample was not designed to do so.

The method was developed using ward-level data from the National Population and Housing Census 2011 based on the new structure of 753 local units. Sampling was conducted by LIRNEasia using the AfterAccess base methodology, with segmentation of wards as described earlier.

The fieldwork was conducted in April-May 2018 by Nielsen Nepal with supervision of all field activities by

LIRNEasia. The data was collected using mobile devices, and uploaded and reviewed on a daily basis, with live monitoring of GPS locations of survey teams.

Raw data collected was then weighted using 2017 national population estimates to correct for over- and under-sampling of certain population sub-groups.

Figure 14. Nepal sample locations based on GPS coordinates recorded during fieldwork

Asia Report 3.0

19

The desired level of accuracy was set to a confidence level of 95% and an absolute precision (relative margin of error) of 5%. The population proportion p was set conservatively to 0.5, which yields the largest sample size. The minimum sample size n was determined by the following equation (Rea & Parker, 1997)3:

Inserting the parameters for the survey yields the minimum sample size for simple random sampling. Depending on the sampling method for the survey, the minimum sample size was multiplied by the design effect variable.

In the absence of empirical data from previous surveys that would have suggested a different value, the default value of 1.5 was chosen for the design effect in all countries except India, where 2 was chosen to account for the additional levels of sample selection. This yielded a minimum sample size of 768 per country for households and individuals. The actual sample size for countries was increased to larger than the minimum requirement to compensate for clustering effects and allow for urban-rural and gender-wise disaggregation of data. In the Asian survey countries, the sample size was increased to 2,000, with the exception of India where the sample size was further increased to 5,000, to ensure precision was maintained with the additional steps in sampling.

3 Rea, L. and Parker, R. (1997): Designing and Conducting Survey Research – A Comprehensive Guide, Jossey-Bass Publishers, San Francisco: Jossey-Bass.

Weighting of dataTwo weights were constructed, one for households, and one for individuals. The weights are based on the inverse selection probabilities. The weights gross up the data to national level.

Wheren = Minimum sample sizeZ = Z - value for 0.05 level of significanceC p = Confidence levelp = Population proportion

( () )Z a 1.96

C p 0.05

p ( 1 - p) 0.5( 1 - 0.5 )n = = = 3842 2

Sample size determination

AfterAccess

20

4 https://data.worldbank.org/indicator/NY.GNP.PCAP.PP.KD, accessed 12 February 2019.

Country Rank GNI per capita, PPP (constant 2011 international $)

Argentina 1 18,444.47

Colombia 2 12,850.37

South Africa 3 11,922.86

Peru 4 11,773.43

Paraguay 5 11,548.06

Sri Lanka 6 11,347.92

Ecuador 7 10,318.93

Guatemala 8 7,284.24

India 9 6,359.19

Pakistan 10 5,311.22

Nigeria 11 5,202.66

Ghana 12 3901.60

Myanmar 13 3,897.59

Bangladesh 14 3,676.92

Cambodia 15 3,419.63

Lesotho 16 3,151.28

Senegal 17 3,054.88

Kenya 18 2,961.94

Tanzania 19 2,556.66

Nepal 20 2,487.93

Rwanda 21 1,814.14

Uganda 22 1,658.20

Mozambique 23 1,100.37

Table 1. GNI per capita of survey countries, PPP

The ordering of the survey countries presented in each chart and table of this report is based on descending GNI per capita (purchasing power parity or PPP terms), at constant 2011 international dollars. This is based on World Bank data for the year 2017 4, as shown in Table 1.

Each data table or graph is accompanied by the relevant survey question and a table of sample bases for each tabulation. Where the number of respondents is low, the base is given in red, and where the number of respondents is insufficient for interpretation, the data is excluded.

Since the Myanmar survey was conducted earlier than the other countries (2016), an early version of the AfterAccess questionnaire was used. Where comparable data is available for Myanmar, it is included in the relevant figures or tables.

Note on reading this report

Asia Report 3.0

21

Connectivity

AfterAccess

22

Mobile phone ownership

Mobile phone ownership ranged from a low of 40% of 15-65-year-olds in Mozambique, the poorest country among the countries surveyed, to 91% of the same age group in Argentina, the richest of the countries surveyed (Figure 15). Most of the Asian countries surveyed were clustered in the region of two-thirds of their 15-65 population owning a mobile phone (i.e.: having an active mobile SIM and device). Most notably among the Asian survey countries, Nepal, the poorest, had mobile penetration of 72% of the 15-65 population at the time of survey – on par with Bangladesh (74%) and just a little behind Sri Lanka (78%) which was the richest of the Asian survey countries.

In all countries, desktop and laptop computer ownership lagged far behind mobile phone ownership. With the exception of the richer Latin American countries, penetration of desktop and laptop computers did not exceed 15% of the 15-65 population. Among the Asian survey countries, only Sri Lanka

(12%) exceeded 6%, but considering its income level, could have been expected to perform better.

The majority of countries had an urban-rural gap in mobile phone ownership, with rural areas of the country lagging behind (Figure 16). Of the Asian countries surveyed, Pakistan, Bangladesh and Sri Lanka had very small gaps, indicating that rural dwellers were almost as likely to own a mobile phone as urban dwellers. The other Asian countries had considerable gaps of 15-22%, meaning that, in these countries, rural dwellers were between 15 and 22 percent less likely to own a mobile phone than urban dwellers.

The gender gap in mobile ownership was highest in India, with women 46% less likely to own a mobile phone than men (Figure 17). Among the other Asian countries surveyed, Pakistan (37%) and Bangladesh (34%) were followed by Myanmar (28%), Cambodia (20%) and Nepal (19%). Interestingly, Sri Lanka’s gender gap

(17%) was much higher than that of its income peers and more in line with its geographical peers. Unlike the more developed, higher-penetration countries of Latin America and South Africa (where in some cases, women were slightly more likely to own a mobile phone than men), gender inequality in mobile phone ownership remained a problem in much of Asia and Africa.

When the data were disaggregated by income within the survey countries (Figure 18), it was clear that those with incomes above the average income of those in the sample had higher levels of mobile phone ownership and vice versa, insofar as they earned something at all.

Asia Report 3.0

23

Q1 Do you own a mobile phone?Q2 How many active SIM cards do you have (SIM cards that you used in the last 30 days)?Q3 Do you own a personal desktop computer or laptop?

Base Arg

entin

a

Colo

mbi

a

Sout

h A

frica

Peru

Para

guay

Sri L

anka

Ecua

dor

Gua

tem

ala

Indi

a

Paki

stan

Nig

eria

Gha

na

Mya

nmar

Ban

glad

esh

Cam

bodi

a

Leso

tho

Sene

gal

Keny

a

Tanz

ania

Nep

al

Rw

anda

Uga

nda

Moz

ambi

que

All respondents 1,240 1,425 1,610 1,478 1,357 2,017 1,420 1,407 5,069 2,002 1,706 1,145 7,204 2,020 2,123 1,844 1,181 1,179 1,102 2,008 1,118 1,757 1,091

84% 89

%43

%

89% 92

%21

%

Arg

entin

a91

%

61%

84%

75%

74%

88%

72%

50% 50

%

40%

85%

68%

61%

75%

92%

61%

88%

87%

82%

83%

76%

89%

73%

53% 56

%

44%

89%

70%

67%

80%

48%

42%

4%

79% 83

%9%

28%

35%

78% 83

%12

%

5% 5%

2% 3%

87%

87%

6%

2%

15%

6%

2%

66%

61%

4%

13%

Indi

a

Sene

gal

Para

guay

Mya

nmar

Rw

anda

Sout

h A

frica

Nig

eria

Tanz

ania

Ecua

dor

Cam

bodi

a

Moz

ambi

que

Colo

mbi

a

Paki

stan

Keny

a

Sri L

anka

Ban

glad

esh

Uga

nda

Peru

Gha

na

Nep

al

Gua

tem

ala

Leso

tho

Desktop/ laptopSIM CardMobile

65%

71%

8%

57%

58%

2%

Mobile phone, SIM card and desktop/laptop ownership (% of population aged 15-65)

Figure 15.

6%

AfterAccess

24

53%

Urban-rural gaps in mobile phone ownership (% of population aged 15-65)

90% 93

%

88%

84%

90%

71%

79%

78% 81

%

88%

76%

63% 65

%

55%

100

%

71%

81% 83

% 86%

55%

72%

64%

77%

65%

46%

45%

33%

Sri L

anka

84%

77%

Ecua

dor

86%

79%

Para

guay

93%

83%

Nig

eria

79%

54%

Paki

stan

59%

56%

Gha

na84

%63

%

Keny

a93

%85

%

Tanz

ania

76%

53%

Sene

gal

85%

74%

GapRuralUrban

Arg

entin

a*

Indi

a

Mya

nmar

Rw

anda

Sout

h A

frica

Cam

bodi

a

Moz

ambi

que

Colo

mbi

a

Ban

glad

esh

Uga

nda

Peru

Nep

al

Gua

tem

ala

Leso

tho

Base Arg

entin

a

Colo

mbi

a

Sout

h A

frica

Peru

Para

guay

Sri L

anka

Ecua

dor

Gua

tem

ala

Indi

a

Paki

stan

Nig

eria

Gha

na

Mya

nmar

Ban

glad

esh

Cam

bodi

a

Leso

tho

Sene

gal

Keny

a

Tanz

ania

Nep

al

Rw

anda

Uga

nda

Moz

ambi

que

All respondents(Urban) 1,208 986 1,050 1,178 824 803 932 550 2,200 793 1,147 721 3,477 808 897 844 790 727 720 940 711 1,024 718

All respondents(Rural) 32 439 765 300 533 1,214 488 857 2,869 1,209 661 479 3,727 1,212 1,226 1,000 391 481 480 538 500 733 453

Q Do you own a mobile phone?

-11%

24%

8%

1%4%

22% 20%

7%

21%

13% 15%

27%31%

41%

Urban-rural gap in mobilephone ownership %

Urban mobile phone owners(% of urban population)

Urban mobile phone owners(% of urban population)= ÷( )- Rural mobile phone owners

(% of rural population)

* Low rural base

30%

9%13%

25%31%

5%8%9%11%

Figure 16.

Asia Report 3.0

25

Gender gaps in mobile phone ownership (% of population aged 15-65)

Base Arg

entin

a

Colo

mbi

a

Sout

h A

frica

Peru

Para

guay

Sri L

anka

Ecua

dor

Gua

tem

ala

Indi

a

Paki

stan

Nig

eria

Gha

na

Mya

nmar

Ban

glad

esh

Cam

bodi

a

Leso

tho

Sene

gal

Keny

a

Tanz

ania

Nep

al

Rw

anda

Uga

nda

Moz

ambi

que

All respondents(Male) 478 487 795 508 879 834 511 656 2,478 1,060 912 547 3,818 1,092 735 515 603 544 531 912 556 848 527

All respondents(Female) 762 938 1,020 970 478 1,183 909 751 2,591 942 896 653 3,386 928 1,388 1,329 578 664 669 1,096 655 909 644

Q Do you own a mobile phone?

GapFemaleMale

Gender gap in mobilephone ownership %

Male mobile phone owners(% of male population)= ÷( )- Female mobile phone owners

(% of female population)Male mobile phone owners

(% of male population)

90%

75%

83% 87

% 91%

79% 82

%

72%

87%

78% 83

%

80%

62%

59%

51%

91%

75%

87%

82%

Sri L

anka

86%

72%

Ecua

dor

84%

83%

84%

43%

Nig

eria

71%

58%

Paki

stan

68%

43%

69%

52%

58% 62

%

81%

Keny

a92

%83

%

Tanz

ania

66%

57%

Sene

gal

83%

75%

65%

39% 42%

32%

Arg

entin

a

Indi

a

Mya

nmar

Rw

anda

Sout

h A

frica

Cam

bodi

a

Moz

ambi

que

Colo

mbi

a

Ban

glad

esh

Uga

nda

Peru

Gha

na

Nep

al

Gua

tem

ala

Leso

tho

-1% 0% -4%

7% 8%16%

28%34%

20%

3%

19%

37%29%

37%

10%15%

9%

46%

18%

37%

17%

0%

Para

guay

5%

92%

88%

Figure 17.

AfterAccess

26

Base Arg

entin

a

Sout

h A

frica

Peru

Para

guay

Sri L

anka

Ecua

dor

Gua

tem

ala

Indi

a

Paki

stan

Nig

eria

Gha

na

Ban

glad

esh

Cam

bodi

a

Leso

tho

Sene

gal

Keny

a

Tanz

ania

Nep

al

Rw

anda

Uga

nda

Moz

ambi

que

All respondents 1,240 1,610 1,478 1,357 2,017 1,420 1,407 5,069 2,002 1,706 1,145 2,020 2,123 1,844 1,181 1,179 1,102 2,008 1,118 1,757 1,091

Q Do you own a mobile phone?

Arg

entin

a

Indi

a

Rw

anda

Sout

h A

frica

Cam

bodi

a

Moz

ambi

que

Ban

glad

esh

Uga

nda

Peru

Gha

na

Nep

al

Gua

tem

ala

Leso

tho

Above average income Below average income Zero income

94%

95%

91% 94

%

78%

92%

89%

81%

95%

85% 90

%

79%

70%

87%

84%

81% 83

%

56%

72%

73%

62%

79% 83

%

49%

47%

40%

91%

76%

76%

Sri L

anka

89%

71%

69%

Ecua

dor

91%

86% 88

%

Para

guay

94%

85% 87

%

83%

49%

Nig

eria

83%

56%

0%

Paki

stan

76%

69%

0%

54%

44%

74%

Keny

a97

%85

%0

%

Tanz

ania

90%

55%

Sene

gal

82%

78%

0%

65%

28%

4%

27%

24%

15%

Mobile phone ownership by income (% of population aged 15-65)

Figure 18.

Asia Report 3.0

27

Type of handset

In the Asian countries surveyed, mobile phone markets comprised predominantly basic handsets with no or limited Internet capability – even in richer Sri Lanka (Figure 19). A basic phone was defined as one on which only calls and texting were possible, and a feature phone as having additional capabilities for multi-media (e.g.: photos, music, etc.) and Internet. A smartphone (also known as a touchphone in Myanmar) was defined as one which used an operating system such as Android, iOS, etc. through which third party ‘apps’ could be run on it, usually with a touch screen (covering 75% or more of its front area).

India performed worst on this aspect (market share with Internet-enabled handsets, i.e.: smart- or feature-phones). Overall, among the Asian survey countries, India’s mobile phone market had the lowest percentage-

share of Internet-enabled mobile phones, though Pakistan had the lowest smartphone penetration. Despite Pakistan’s (and Bangladesh’s) low smartphone penetration, it still had a significant feature phone segment, enabling users to access the Internet. Interestingly, Cambodia and Nepal – the lowest-income Asian countries studied – had smartphone penetration higher (close to half the market in both cases) than the richer Asian countries studied, including Sri Lanka, the richest. Similarly, Myanmar (whose mobile market was liberalized most recently, in 2013) had a smartphone penetration of 78% of the market by 2016 when the survey was conducted.5

As expected, incidence of smartphone ownership among urban respondents was higher than among rural respondents (Figure 20), a feature of disparities in income as well

as, perhaps, device availability and perceived relevance between urban and rural. The difference in smartphone ownership between urban and rural respondents in the Asian countries surveyed appears to be largest in India and Pakistan.

The gender gap was not as large with respect to device type (Figure 21) as with mobile phone ownership, noted earlier. It appears that the bigger hurdle is for women to get connected (become mobile phone owners). Thereafter, they are sometimes as likely as men to get an Internet-enabled phone.

5 The Myanmar data is not depicted in the relevant graph due to slight differences in the survey question responses. It can not be directly compared with that from the other survey countries. For further information, the Myanmar report can be viewed at: bit.ly/LIRNEasiaMyanmar2016.

AfterAccess

28

Base Arg

entin

a

Colo

mbi

a

Sout

h A

frica

Peru

Para

guay

Sri L

anka

Ecua

dor

Gua

tem

ala

Indi

a

Paki

stan

Nig

eria

Gha

na

Ban

glad

esh

Cam

bodi

a

Sene

gal

Keny

a

Tanz

ania

Nep

al

Rw

anda

Uga

nda

Moz

ambi

que

Mobile phone owners 1,116 1,297 1,398 1,234 1,209 1,609 1,191 1,214 3,252 1,208 1,123 901 1,531 1,526 969 1,054 761 1,478 635 1,031 632

Q What type of mobile phone is it?

Smartphone Feature phone Basic phone

Arg

entin

a

Indi

a

Rw

anda

Sout

h A

frica

Colo

mbi

a

Cam

bodi

a

Moz

ambi

que

Ban

glad

esh

Uga

nda

Peru

Gha

na

Nep

al

Gua

tem

ala

78%

63%

58%

61%

57%

28%

35%

24%

48% 52

%

9%

17%

18%

10%

10%

8%

17%

14%

16% 14

%

37%

10% 8%

25% 12%

5%

12%

28%

34% 22

%

Sri L

anka

47%

7%46

%

Ecua

dor

76%

12%

12%

Para

guay

65%

5%30

%

30%

55%

Nig

eria

23%

46%

31%

Paki

stan

22%

25%

53%

52% 40

%

42%

Keny

a28

%14

%58

%

Tanz

ania

23%

11%

66%

Sene

gal

31%

12%

58%

40%

66%

71%

77%

Handset type owned (% of population aged 15-65)

Figure 19.

Asia Report 3.0

29

Base Arg

entin

a

Colo

mbi

a

Sout

h A

frica

Peru

Para

guay

Sri L

anka

Ecua

dor

Gua

tem

ala

Indi

a

Paki

stan

Nig

eria

Gha

na

Ban

glad

esh

Cam

bodi

a

Sene

gal

Keny

a

Tanz

ania

Nep

al

Rw

anda

Uga

nda

Moz

ambi

que

Mobile phone owners 1,116 1,297 1,398 1,234 1,209 1,609 1,191 1,214 3,252 1,208 1,123 901 1,531 1,526 969 1,054 761 1,478 635 1,031 632

Q What type of mobile phone is it?

Figure 20.

Type of handset owned, by location (% of population aged 15-65)

Arg

entin

a

Rw

anda

Colo

mbi

a

Moz

ambi

que

Uga

nda

Nep

al

Keny

a

Tanz

ania

Sene

gal

12%

U R

U R

U R U R U R U R U R U R U R U R U R U R U R U R U R U R U R U R U R U R U R U R

8%

21%

30% 27

%47

%

16%

42%

22%

46% 40

%47

%

11%

13%

28%

31%

45%

62%

44%

58%

26%

35%

45%

62%

33%

42% 35

%44

%

49%

66%

32%

68%

51%

77%

38%

44%

50%

72%

49%

80%

60%

91%

10% 9%

5%11

% 8%10

%

17%

18%

5%4%

6%7%

12%

12%

13%

14%

15%

17%

23%

26%

44%

48%

14%

13%

40%

35%

5%12

%

14%

9%

13%

15%

9%12

%

8%

24%

14%

7%

9%

25%

12%

4%5%

78% 83

%

73%

59% 65

%43

%

67%

39%

73%

50%

53%

45%

77%

75%

59%

55%

40%

20%

33%

16%

30%

17%

41%

25%

27%

22%

60%

44%

36%

25%

55%

17%

40%

10%

54%

47%

26%

3%

37%

8%

32%

Smartphone Feature phone Basic phone

Urban Rural

Indi

a

Cam

bodi

a

Ban

glad

esh

Peru

Gha

na

Gua

tem

ala

Sri L

anka

Ecua

dor

Para

guay

Nig

eria

Paki

stan

Sout

h A

frica

AfterAccess

30

Figure 21.

Base Arg

entin

a

Colo

mbi

a

Sout

h A

frica

Peru

Para

guay

Sri L

anka

Ecua

dor

Gua

tem

ala

Indi

a

Paki

stan

Nig

eria

Gha

na

Ban

glad

esh

Cam

bodi

a

Sene

gal

Keny

a

Tanz

ania

Nep

al

Rw

anda

Uga

nda

Moz

ambi

que

Mobile phone owners 1,116 1,297 1,398 1,234 1,209 1,609 1,191 1,214 3,252 1,208 1,123 901 1,531 1,526 969 1,054 761 1,478 635 1,031 632

Q What type of mobile phone is it?

Type of handset owned, by gender (% of population aged 15-65)

Arg

entin

a

Rw

anda

Moz

ambi

que

Uga

nda

Nep

al

Colo

mbi

a

Keny

a

Tanz

ania

Sene

gal

M F

Smartphone Feature phone Basic phone

Male Female

M F M F M F M F M F M F M F M F M F M F M F M F M F M F M F M F M F M F M F M F M F

76%

60%

63%

64%

63%

52%

81%

65%

31%

20%

26% 35

%

25%

52%

37%

32%

25%

56%

10% 16

% 19%

10%

12% 7%

15%

3%

6%

10%

13%

16%

28%

46%

13% 38

%

12%

9% 17%

11%

6%

21%

11%

5%

14%

28%

30% 20

%

34%

42%

9%

22%

53%

52%

28%

52% 36

%

36%

55%

51%

64%

38%

69%

73%

76%

81%

64%

54% 59

% 66%

42%

73%

49%

24% 27

%

20% 34

%

20%

46%

24%

25%

21%

48%

9%

18%

16%

10%

8%

9%

18%

5%

8%

13%

14%

17% 20

% 46%

15%

34%

8%

15%

12%

11%

11%

29% 15%

6%

9%

28%

37%

22%

28%

50%

14%

37%

60%

53%

34%

51% 46

%

46%

61%

63%

69%

41%

61%

67%

79%

Sout

h A

frica

Indi

a

Cam

bodi

a

Ban

glad

esh

Peru

Gha

na

Gua

tem

ala

Sri L

anka

Ecua

dor

Para

guay

Nig

eria

Paki

stan

Asia Report 3.0

31

New adoptersMost of the Asian countries surveyed connected approximately 30% of their current subscribers in the past three years (Table 2), predominantly from among rural dwellers (Table 3) and women (Table 4).

Country Number of years since becoming a mobile phone owner

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 15+

Argentina 4% 3% 4% 3% 7% 7% 4% 6% 3% 23% 3% 6% 3% 2% 9% 15%

Colombia 11% 10% 7% 5% 10% 9% 5% 7% 3% 10% 1% 6% 1% 1% 7% 9%

South Africa 6% 6% 5% 5% 4% 4% 7% 4% 4% 4% 4% 5% 4% 4% 3% 30%

Peru 5% 7% 8% 6% 13% 6% 6% 7% 4% 15% 2% 4% 2% 1% 7% 8%

Paraguay 6% 6% 8% 6% 12% 6% 6% 6% 2% 15% 1% 5% 2% 1% 9% 11%

Sri Lanka 15% 8% 9% 7% 6% 5% 3% 7% 3% 9% 3% 3% 4% 2% 3% 12%

Ecuador 6% 7% 9% 5% 10% 6% 3% 7% 2% 15% 2% 5% 2% 2% 7% 11%

Guatemala 13% 11% 11% 5% 11% 6% 3% 7% 2% 12% 1% 3% 0% 1% 5% 7%

India 15% 14% 10% 6% 12% 6% 9% 3% 4% 6% 2% 4% 1% 1% 2% 4%

Pakistan 9% 10% 8% 3% 6% 6% 10% 4% 8% 3% 3% 7% 3% 2% 3% 13%

Nigeria 7% 5% 7% 5% 9% 4% 7% 5% 8% 11% 6% 8% 6% 3% 4% 6%

Ghana 14% 9% 7% 9% 6% 6% 9% 6% 5% 9% 3% 4% 3% 1% 4% 4%

Bangladesh 8% 12% 10% 7% 11% 6% 11% 6% 8% 5% 3% 5% 1% 1% 1% 4%

Cambodia 12% 6% 9% 7% 11% 4% 8% 3% 3% 12% 3% 5% 2% 3% 4% 8%

Senegal 2% 8% 12% 8% 6% 7% 7% 4% 7% 5% 6% 5% 4% 4% 3% 14%

Kenya 11% 9% 8% 8% 8% 4% 10% 9% 5% 6% 3% 4% 3% 2% 3% 7%

Tanzania 11% 7% 7% 7% 6% 7% 12% 10% 7% 6% 4% 6% 3% 2% 1% 5%

Nepal 5% 9% 13% 9% 15% 9% 10% 5% 3% 9% 4% 2% 2% 1% 2% 2%

Rwanda 18% 11% 11% 5% 15% 5% 3% 7% 3% 9% 2% 2% 1% 1% 3% 3%

Uganda 7% 11% 11% 10% 7% 7% 5% 3% 7% 4% 3% 4% 1% 4% 2% 12%

Mozambique 12% 12% 8% 8% 8% 9% 6% 3% 4% 6% 3% 7% 3% 2% 4% 5%

Number of years since becoming a mobile phone owner (% of mobile phone owners aged 15-65)

Table 2.

AfterAccess

32

Arg

entin

a

Colo

mbi

a

Sout

h A

frica

Peru

Para

guay

Sri L

anka

Ecua

dor

Gua

tem

ala

Indi

a

Paki

stan

Nig

eria

Gha

na

Ban

glad

esh

Cam

bodi

a

Sene

gal

Keny

a

Tanz

ania

Nep

al

Rw

anda

Uga

nda

Moz

ambi

que

Urb

an

Rur

al

Urb

an

Rur

al

Urb

an

Rur

al

Urb

an

Rur

al

Urb

an

Rur

al

Urb

an

Rur

al

Urb

an

Rur

al

Urb

an

Rur

al

Urb

an

Rur

al

Urb

an

Rur

al

Urb

an

Rur

al

Urb

an

Rur

al

Urb

an

Rur

al

Urb

an

Rur

al

Urb

an

Rur

al

Urb

an

Rur

al

Urb

an

Rur

al

Urb

an

Rur

al

Urb

an

Rur

al

Urb

an

Rur

al

Urb

an

Rur

al

< 1 4 6 6 13 6 6 4 10 6 7 16 15 5 7 11 15 13 17 7 10 8 6 6 9 10 20 9 13 2 1 5 13 5 15 5 6 15 19 6 7 5 18

2-5 17 19 16 36 17 29 30 43 29 34 33 30 32 31 41 37 41 42 26 29 22 30 36 43 31 33 27 35 32 35 29 34 22 32 42 54 40 44 37 40 28 41

6-10 43 50 34 33 24 22 39 35 36 36 22 28 34 35 29 32 28 29 31 31 35 34 38 35 36 33 31 29 28 31 35 34 44 38 38 31 28 26 27 27 36 23

11-15 22 6 25 13 22 16 17 10 17 17 17 14 16 18 12 11 13 8 20 18 28 25 15 9 18 11 21 16 21 22 21 13 23 10 12 9 10 9 15 14 22 16

15+ 15 19 19 5 31 28 9 2 13 7 13 12 12 9 7 6 5 4 16 12 7 5 5 4 6 2 11 7 18 10 10 6 6 5 2 1 6 2 15 11 9 2

Arg

entin

a

Colo

mbi

a

Sout

h A

frica

Peru

Para

guay

Sri L

anka

Ecua

dor

Gua

tem

ala

Indi

a

Paki

stan

Nig

eria

Gha

na

Ban

glad

esh

Cam

bodi

a

Sene

gal

Keny

a

Tanz

ania

Nep

al

Rw

anda

Uga

nda

Moz

ambi

que

Mal

e

Fem

ale

Mal

e

Fem

ale

Mal

e

Fem

ale

Mal

e

Fem

ale

Mal

e

Fem

ale

Mal

e

Fem

ale

Mal

e

Fem

ale

Mal

e

Fem

ale

Mal

e

Fem

ale

Mal

e

Fem

ale

Mal

e

Fem

ale

Mal

e

Fem

ale

Mal

e

Fem

ale

Mal

e

Fem

ale

Mal

e

Fem

ale

Mal

e

Fem

ale

Mal

e

Fem

ale

Mal

e

Fem

ale

Mal

e

Fem

ale

Mal

e

Fem

ale

Mal

e

Fem

ale

< 1 4 4 11 11 5 6 6 5 5 7 16 15 6 6 12 14 12 21 9 7 7 7 6 12 12 16 8 14 2 1 9 12 7 14 3 8 14 23 8 7 11 14

2-5 16 18 26 33 18 23 25 37 24 34 33 30 28 34 34 43 40 46 26 32 25 27 38 47 30 34 26 38 30 38 31 34 26 29 39 53 40 47 35 46 36 34

6-10 36 49 39 30 23 23 36 39 36 35 22 28 36 33 33 29 31 23 32 29 34 36 37 33 36 34 32 28 29 30 36 33 41 41 42 29 31 22 26 28 27 30

11-15 25 19 10 20 21 19 20 13 18 16 17 14 17 17 12 10 12 7 19 18 28 25 13 7 17 14 22 14 23 18 14 16 17 14 13 9 11 7 16 12 20 17

15+ 19 11 13 6 32 29 12 5 16 8 13 12 14 10 9 5 5 3 13 14 7 5 6 2 5 3 12 5 16 12 9 6 8 2 2 1 5 1 16 8 5 5

Base Arg

entin

a

Colo

mbi

a

Sout

h A

frica

Peru

Para

guay

Sri L

anka

Ecua

dor

Gua

tem

ala

Indi

a

Paki

stan

Nig

eria

Gha

na

Ban

glad

esh

Cam

bodi

a

Sene

gal

Keny

a

Tanz

ania

Nep

al

Rw

anda

Uga

nda

Moz

ambi

que

Mobile phone owners 1,116 1,297 1,398 1,234 1,209 1,609 1,191 1,214 3,252 1,208 1,123 901 1,531 1,526 969 1,054 761 1,478 635 1,031 632

Q When did you get your first ever mobile phone connection (i.e.: working handset and SIM card)? (date converted to time since)

Number of years since becoming a mobile phone owner, by location (% of mobile phone owners aged 15-65)

Number of years since becoming a mobile phone owner, by gender (% of mobile phone owners aged 15-65)

Table 3.

Table 4.

Asia Report 3.0

33

Multiple SIM ownershipMore than one quarter of mobile phone owners in the Asian countries surveyed had more than one active SIM at the time of survey (Table 5). This includes a SIM that had been used in the 30 days preceding survey. The actual number ranged from 34% in Bangladesh to 23% in Pakistan. Though similar if not higher numbers were seen in the African countries surveyed, multiple SIM use was less common in the Latin American countries surveyed.

Multiple SIM ownership was higher among men, urban dwellers (except in Pakistan and Sri Lanka, where the relationship was reversed), those with higher levels of income and those of the younger age groups (below 35 years) (Table 6).

Arg

entin

a

Colo

mbi

a

Sout

h A

frica

Peru

Para

guay

Sri L

anka

Ecua

dor

Gua

tem

ala

Indi

a

Paki

stan

Nig

eria

Gha

na

Mya

nmar

Ban

glad

esh

Cam

bodi

a

Leso

tho

Sene

gal

Keny

a

Tanz

ania

Nep

al

Rw

anda

Uga

nda

Moz

ambi

que

1 93% 79% 80% 83% 89% 68% 89% 87% 74% 77% 48% 57% 73% 66% 71% 68% 68% 80% 59% 72% 58% 56% 76%

2 6% 20% 18% 13% 9% 29% 10% 12% 23% 19% 42% 38% 25% 30% 26% 28% 30% 18% 36% 28% 37% 41% 21%

3 1% 1% 2% 3% 1% 2% 1% 1% 2% 3% 7% 3% 1% 2% 3% 3% 3% 1% 4% 0% 5% 2% 3%

4 0% 0% 0% 0% 0% 1% 0% 0% 1% 1% 3% 1% 1% 1% 1% 0% 0% 0% 1% 0% 0% 0% 0%

5 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 1% 0% 0% 0% 0% 0% 0% 0% 0% 0%

6+ 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 1% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0% 0%

Base Arg

entin

a

Colo

mbi

a

Sout

h A

frica

Peru

Para

guay

Sri L

anka

Ecua

dor

Gua

tem

ala

Indi

a

Paki

stan

Nig

eria

Gha

na

Mya

nmar

Ban

glad

esh

Cam

bodi

a

Leso

tho

Sene

gal

Keny

a

Tanz

ania

Nep

al

Rw

anda

Uga

nda

Moz

ambi

que

Mobile phone owners 1,116 1,297 1,398 1,234 1,209 1,609 1,191 1,214 3,252 1,208 1,123 901 4390 1,531 1,526 1,708 969 1,054 761 1,478 635 1,031 632

Q How many active SIM cards do you have, (SIM cards that you used in the last 30 days)?

Number of active SIM cards (% of population aged 15-65)

Table 5.

AfterAccess

34

Arg

entin

a

Peru

Para

guay

Sri L

anka

Ecua

dor

Gua

tem

ala

Indi

a

Paki

stan

Nig

eria

Gha

na

Ban

glad

esh

Cam

bodi

a

Leso

tho

Sene

gal

Keny

a

Tanz

ania

Nep

al

Rw

anda

Loca

tion Urban 7% 17% 12% 28% 11% 12% 29% 20% 55% 46% 35% 35% 36% 32% 25% 51% 30% 55%

Rural 10% 16% 8% 33% 10% 13% 24% 26% 48% 39% 34% 27% 29% 33% 17% 34% 25% 37%

Gen

der Male 8% 23% 14% 37% 17% 14% 29% 28% 58% 51% 38% 33% 35% 39% 23% 42% 31% 42%

Female 7% 13% 9% 27% 7% 11% 21% 15% 44% 34% 28% 27% 31% 25% 16% 41% 25% 42%

Educ

atio

n Secondary or higher 7% 18% 13% 35% 11% 12% 31% 34% 58% 47% 43% 41% 54% 44% 23% 50% 39% 65%

Primary or none 0% 9% 8% 19% 11% 13% 22% 22% 35% 38% 27% 24% 30% 25% 15% 31% 14% 32%

Inco

me

Above average 12% 26% 16% 35% 20% 16% 30% 24% 61% 50% 38% 37% 47% 44% 41% 51% 34% 50%

Below average 4% 11% 7% 39% 13% 10% 24% 24% 45% 42% 32% 25% 30% 28% 14% 34% 30% 41%

Zero income 4% 11% 7% 23% 10% 9% 23% 19% 0% 0% 31% 26% 22% 0% 0% 40% 26% 32%

Age

15-25 4% 15% 10% 42% 8% 9% 33% 23% 43% 36% 39% 27% 29% 33% 20% 44% 34% 39%

26-35 7% 18% 13% 46% 11% 12% 27% 27% 59% 52% 38% 35% 39% 32% 19% 46% 32% 49%

36-45 13% 18% 10% 32% 13% 17% 25% 22% 58% 46% 33% 33% 38% 39% 22% 42% 24% 49%

46-55 9% 18% 11% 22% 15% 16% 19% 18% 48% 47% 31% 28% 28% 24% 22% 30% 18% 33%

56-65 5% 11% 9% 14% 8% 19% 16% 24% 42% 28% 18% 22% 22% 31% 10% 32% 12% 23%

Base Arg

entin

a

Peru

Para

guay

Sri L

anka

Ecua

dor

Gua

tem

ala

Indi

a

Paki

stan

Nig

eria

Gha

na

Ban

glad

esh

Cam

bodi

a

Leso

tho

Sene

gal

Keny

a

Tanz

ania

Nep

al

Rw

anda

Mobile phone owners 1,116 1,234 1,209 1,609 1,191 1,214 3,252 1,208 1,123 901 1,531 1,526 1,708 969 1,054 761 1,478 635

Q How many active SIM cards do you have, (SIM cards that you used in the last 30 days)?

Multiple SIM use, by location, gender, education level, income level and age group (% of mobile phone owners aged 15-65)

Table 6.

Asia Report 3.0

35

InternetLack of awareness of the Internet was a considerable problem across Asian (and, to a lesser extent, African) countries surveyed (Figure 22). For example, in India just 35% of the population aged 15-65 said they knew what the Internet was, with similar numbers seen in Pakistan and Bangladesh. Surprisingly in Nepal – the poorest of the Asian countries surveyed – 46% of the 15-65 population knew what the Internet was.

The levels of use were even lower than levels of awareness in all the Asian countries surveyed, except Cambodia. This anomaly could be interpreted as a possible difference in the manner in which the questions were asked on the field in Cambodia. Myanmar, Cambodia and Nepal (the poorer three of the Asian countries surveyed) had higher levels of Internet use than the other Asian countries surveyed.

The urban-rural gap in Internet use was large in most countries (even the higher income ones), with rural-dwellers lagging behind as much as 48% in India and 32-42% in Bangladesh, Cambodia, Myanmar and Nepal (Figure 23). Sri Lanka’s urban-rural gap was just 23%.

The gender gap in Internet use was markedly larger in the Asian and

African countries surveyed than the Latin American ones (Figure 24), even in Sri Lanka whose income level is on par with that of the Latin American countries.

Smartphone owners are the highest users of the Internet in the Asian countries surveyed (Figure 25). Unlike in the Latin American countries or some of the African countries surveyed, feature phone owners in the Asian countries barely used the Internet.

Current Internet users were asked if there was anything that limited their use of the Internet. Of the Asian countries surveyed, Nepal, Bangladesh, Sri Lanka and India had more than a quarter of Internet users who stated that their Internet use was not limited by any factors (Figure 26). While the Asian survey countries performed slightly better than the African survey countries in terms of (perceived) cost of data, it was cited as a primary limitation to greater use by 30% of Internet users aged 15-65 in Bangladesh, 29% in Nepal and 25% in Sri Lanka. The speed of the Internet was a major barrier to greater use in Cambodia, with 70% of Internet users aged 15-65 citing it as a primary limiting factor, compared to much lower numbers in the other Asian

countries surveyed. A considerable number of users in Pakistan (50%) cited ‘lack of time’ as a primary limiting factor to more Internet use.

Figure 26 contains responses on main limiting factors in survey countries where respondents were asked to cite the primary factor (single response) while Figure 27 presents the responses from countries where respondents were allowed to name more than one factor if applicable. The question was asked slightly differently in different countries. As such, a wider set of concerns was voiced in the latter set of countries (Figure 27), with issues related to cybersecurity and privacy being cited more commonly in Latin America. In Africa, data cost, speed of the Internet and lack of time were the more common concerns.

As Figure 28 shows, Internet and social media use went hand-in-hand across the countries surveyed. Social media use levels among the 15-65 population were as low as Internet use in the Asian countries surveyed. The following section provides more detail on social media use. Sri Lanka, Cambodia and Nepal appeared to be further ahead in social media use as well as Internet use than their geographic peers. However, Sri Lanka lagged behind its income peers.

AfterAccess

36

A lack of awareness of what the Internet is, was the key barrier to Internet use across survey countries (Figure 29), with more than 60% of current non-users in the Asian countries (as high as 97% in Cambodia) citing this as a main reason for non-use. In the African survey countries, the lack of a device seemed to be a key concern, while in Latin America, the lack of devices was a smaller concern and not knowing how to use the Internet was also an issue.

Surprisingly, there was non-awareness even among current smartphone owners who didn’t use the Internet, across survey countries (Figure 30).

Arg

entin

a

Colo

mbi

a

Sout

h A

frica

Peru

Para

guay

Sri L

anka

Ecua

dor

Gua

tem

ala

Indi

a

Paki

stan

Nig

eria

Gha

na

Mya

nmar

*

Ban

glad

esh

Cam

bodi

a

Leso

tho*

Sene

gal

Keny

a

Tanz

ania

Nep

al

Rw

anda

Uga

nda

Moz

ambi

que

81% 86

%

76%

73%

72%

53%

76%

73%

67%

63%

62%

37%

85%

83%

79%

65%

35%

19%

37%

17%

54%

30%

48%

26% 30

% 33%

13%

11%

36%

36%

52%

31%

55%

26%

48%

15%

46%

34%

23%

9%

32%

15%

41%

10%

Internet awareness Internet use

Internet awareness and use (% of population aged 15-65)

Figure 22.

Base Arg

entin

a

Colo

mbi

a

Sout

h A

frica

Peru

Para

guay

Sri L

anka

Ecua

dor

Gua

tem

ala

Indi

a

Paki

stan

Nig

eria

Gha

na

Mya

nmar

Ban

glad

esh

Cam

bodi

a

Leso

tho

Sene

gal

Keny

a

Tanz

ania

Nep

al

Rw

anda

Uga

nda

Moz

ambi

que

All respondents 1,240 1,425 1,610 1,478 1,357 2,017 1,420 1,407 5,069 2,002 1,706 1,145 7,204 2,020 2,123 1,844 1,181 1,179 1,102 2,008 1,118 1,757 1,091

Q1 Do you know what the Internet is?Q2 Have you ever used the Internet (Gmail, Google, Facebook, email etc.)?

* No data available for Q1

Asia Report 3.0

37

Arg

entin

a*

Colo

mbi

a

Sout

h A

frica

Peru

Para

guay

Sri L

anka

Ecua

dor

Gua

tem

ala

Indi

a

Paki

stan

Nig

eria

Gha

na

Mya

nmar

Ban

glad

esh

Cam

bodi

a

Leso

tho

Sene

gal

Keny

a

Tanz

ania

Nep

al

Rw

anda

Uga

nda

Moz

ambi

que

86% 91

%

90%

69%

61%

40%

81%

49%

74%

44%

45%

35%

85%

80%

70%

61%

27%

14% 18

%16

%

42%

22%

36%

17%

45%

28%

19%

11%

51%

31%

55%

20%

41%

23%

53%

16%

33%

5%

38%

26%

22%

5%

31%

10%

24%

4%

GapRuralUrban

Urban-rural gap in Internet use (% of population aged 15-65)

Figure 23.

Base Arg

entin

a

Colo

mbi

a

Sout

h A

frica

Peru

Para

guay

Sri L

anka

Ecua

dor

Gua

tem

ala

Indi

a

Paki

stan

Nig

eria

Gha

na

Mya

nmar

Ban

glad

esh

Cam

bodi

a

Leso

tho

Sene

gal

Keny

a

Tanz

ania

Nep

al

Rw

anda

Uga

nda

Moz

ambi

que

All respondents(Urban) 1,208 986 1,050 1,178 824 803 932 550 2,200 793 1,147 721 3,477 808 897 844 790 727 720 940 711 1,024 718

All respondents(Rural) 32 439 765 300 533 1,214 488 857 2,869 1,209 661 479 3,727 1,212 1,226 1,000 391 481 480 538 500 733 453

Q Have you ever used the Internet (Gmail, Google, Facebook, email etc.)?

Urban-rural gap in Internet use %

Urban Internet users(% of urban population) = ÷( )- Rural Internet users

(% of rural population) Urban Internet users

(% of urban population)

* Low rural base

-6%

23%35% 40% 40%

23%

6%13%

48%

13%

49% 54%38% 42% 40%

63%

45%

69%84%

32%

77%67%

85%

AfterAccess

38

87%

71%

57%

81%

63%

45%

89%

75%

26%

21%

39%

33%

45%

40%

35%

32%

18%

41%

13% 17

%

14%

86%

75%

50%

69%

64%

30%

80%

57%

11% 12%

21% 22

%

7%

30% 35

%

27%

22%

12%

27%

5%

13%

7%

1%-6%

12% 14%-1%

34%

10%23%

57%43% 46%

34%

62%

33% 34%

13% 21%31% 32% 33%

62%

25%

50%

Arg

entin

a

Colo

mbi

a

Sout

h A

frica

Peru

Para

guay

Sri L

anka

Ecua

dor

Gua

tem

ala

Indi

a

Paki

stan

Nig

eria

Gha

na

Mya

nmar

Ban

glad

esh

Cam

bodi

a

Leso

tho

Sene

gal

Keny

a

Tanz

ania

Nep

al

Rw

anda

Uga

nda

Moz

ambi

que

Gender gap in Internet use (% of population aged 15-65)

Figure 24.

Base Arg

entin

a

Colo

mbi

a

Sout

h A

frica

Peru

Para

guay

Sri L

anka

Ecua

dor

Gua

tem

ala

Indi

a

Paki

stan

Nig

eria

Gha

na

Mya

nmar

Ban

glad

esh

Cam

bodi

a

Leso

tho

Sene

gal

Keny

a

Tanz

ania

Nep

al

Rw

anda

Uga

nda

Moz

ambi

que

All respondents(Male) 478 487 795 508 879 843 511 656 2,478 1,060 912 547 3,818 1,092 735 515 603 544 531 912 556 848 527

All respondents(Female) 762 938 1,020 970 478 1,183 909 751 2,591 942 896 653 3,386 928 1,388 1,329 578 664 669 1,096 655 909 644

Q Have you ever used the Internet (Gmail, Google, Facebook, email etc.)?

Gender gap in Internet use %

Male Internet users(% of male population) = ÷( )- Female Internet users

(% of female population) Male Internet users

(% of male population)

GapFemaleMale

7%

24%

18%

37%

Asia Report 3.0

39

Internet use, by type of handset owned (% of mobile phone owners aged 15-65) Figure 25.

Base Arg

entin

a

Colo

mbi

a

Sout

h A

frica

Peru

Para

guay

Sri L

anka

Ecua

dor

Gua

tem

ala

Indi

a

Paki

stan

Nig

eria

Gha

na

Ban

glad

esh

Cam

bodi

a

Sene

gal

Keny

a

Tanz

ania

Nep

al

Rw

anda

Uga

nda

Moz

ambi

que

Mobile phone owners 1,116 1,297 1,398 1,234 1,209 1,609 1,191 1,214 3,252 1,208 1,123 901 1,531 1,526 969 1,054 761 1,478 635 1,031 632

Q Have you ever used the Internet (Gmail, Google, Facebook, email etc.)?

Smartphone Feature phone Basic phone

Arg

entin

a

Colo

mbi

a

Sout

h A

frica

Peru

Para

guay

Sri L

anka

Ecua

dor

Gua

tem

ala

Indi

a

Paki

stan

Nig

eria

Gha

na

Ban

glad

esh

Cam

bodi

a

Sene

gal

Keny

a

Tanz

ania

Nep

al

Rw

anda

Uga

nda

Moz

ambi

que

40%

83%

97%

39%

83%

98%

14%

53%

84%

33%

68%

95%

7%73

%98

%

10%15

%80

%

37%

71%

97%

22%

88%92

%

5%14

%78

%

12%

22%

65%

16%

39%

88%

6%26

%82

%

1%5%

64%

11%

10%

88%

4%74

%90

%

2%30

%86

%

3%19

%87

%

3%25

%82

%

5%15

%87

%

5%45

%86

%

4%58

%89

%

AfterAccess

40

29%

25%

25%

5%

4%4%

4%2%

1%1%

Sri Lanka

27%

18%

21%

22%

2%2%

3%2%

1%2%

1%

India

22%

18%

49%

3%1%

3%2%

2%1%

Pakistan

18%

33%

16%

17%

9% 3%1%

2%

Nigeria

12%

52%

21%

8% 4%1%

1%1%

Ghana

33%

30%

19%

15% 1%

1%1%

1%

Bangladesh

11%

5%7%

70%

2%4%

1%

Cambodia26

%24

%40

%6%

2%1%

1%

Senegal

17%

45%

20%

12%

2%1%

1%2%

1%

Kenya

40%

29%

18%

5%

1%1%

1%3%

1%

Nepal

Main limitation to Internet use (% of Internet users aged 15-65)

Figure 26.

Base Sri L

anka

Indi

a

Paki

stan

Nig

eria

Gha

na

Ban

glad

esh

Cam

bodi

a

Sene

gal

Keny

a

Nep

al

Internet users 739 919 427 529 311 266 804 393 440 692

Q What is the main limitation to your use of the Internet?*

* Single-response question

Nothing, no limitation

Speed of Internet

Cost of data

Lack of time

Other

Not allowed (by family, spouse, parents) to use it more

Privacy concerns

No Internet connection in the area

I find it difficult to use

Lack of content in my language

Worried about getting virus/malware

Asia Report 3.0

41

AfterAccess

42

Limitations to Internet use (% of Internet users aged 15-65)

Figure 27.

Worried about getting virus/malware

Worried about surveillance/privacy invasion

The Internet is too expensive to use

The Internet is very slow

Lack of time

Lack of interesting content for me

None

Few people to communicate with via the Internet

Lack of local language content

I find it difficult to use

Someone restricting use (e.g.: family, spouse, parents)

Other

72%

68%

58%

51%

40%

31%

30%

27%

24%

9% 9%1%

Argentina

90%

83%

66%

58%

41%

21%

33%

30%

30%

19%

16%

Colombia

4% 3%47

%24

%10

%8%

16%

6%3% 2% 3%

6%

South Africa

84%

78%

53%

59%

58%

48%

9%39

% 42%

18%

18%

4%

Peru

53%

65%

56%

61%

29%

27%

19%

30%

30%

15%

8%

Paraguay

87%

83%

49%

54% 56

%52

%2%

43% 45

%17

%25

%1%

Ecuador

0%

0%

Asia Report 3.0

43

Guatemala

Base Arg

entin

a

Colo

mbi

a

Sout

h A

frica

Peru

Para

guay

Ecua

dor

Gua

tem

ala

Leso

tho

Tanz

ania

Rw

anda

Uga

nda

Moz

ambi

que

Internet users 1,006 1,192 829 1,120 886 1,165 1,104 652 266 172 346 238

Q I’m going to read some phrases that other people have mentioned as limitations to the use of Internet. For each one, please, tell me if you consider it a limitation or not.*

* Multiple-response question

84%

84%

52% 54

%49

%38

%24

%41

%38

%21

%26

%

17%

11%

40%

21%

4%23

%6%

4%

26%

1%

0%

0%

0%

0%

0%

0%1%

41%

29%

26%

5% 6% 6%4% 4%

1%6%

10%

8%55

%19

%55

%9%

24%

14%

6%3% 3%

5%65

%43

%33

%8% 8%

19%

6%11

%10

%5%

9%2%

43%

37%

12%

6%11

% 12%

6% 5% 5% 6%

Lesotho Tanzania Rwanda Uganda Mozambique

AfterAccess

44

30%

21%

Arg

entin

a

Indi

a

Leso

tho

Para

guay

Rw

anda

Sout

h A

frica

Colo

mbi

a

Nig

eria

Tanz

ania

Ecua

dor

Cam

bodi

a

Moz

ambi

que

Paki

stan

Keny

a

Sene

gal

Sri L

anka

Ban

glad

esh

Uga

nda

Peru

Mya

nmar

Gha

na

Nep

al

Gua

tem

ala

Internet and social media use (% of population aged 15-65)

Base Arg

entin

a

Colo

mbi

a

Sout

h A

frica

Peru

Para

guay

Sri L

anka

Ecua

dor

Gua

tem

ala

Indi

a

Paki

stan

Nig

eria

Gha

na

Mya

nmar

Ban

glad

esh

Cam

bodi

a

Leso

tho

Sene

gal

Keny

a

Tanz

ania

Nep

al

Rw

anda

Uga

nda

Moz

ambi

que

All respondents 1,240 1,425 1,610 1,478 1,357 2,017 1,420 1,407 5,069 2,002 1,706 1,145 7,204 2,020 2,123 1,844 1,181 1,179 1,102 2,008 1,118 1,757 1,091

Internet use Social media use

86%

73%

53%

73%

63%

37%

83%

65%

19%

17%

30%

26%

36%

36%

31%

26%

15%

34%

9%

15%

10%

82%

80%

45%

67%

62%

29%

80%

62%

15%

14%

30%

29%

30%

31%

31%

26%

13%

32%

6%

13%

9%

Q1 Have you ever used the Internet (Gmail, Google, Facebook, email etc.)?Q2 Do you use social media like Facebook, WhatsApp, Twitter etc.?

Figure 28.

13%

13%

Asia Report 3.0

45

Base Arg

entin

a

Colo

mbi

a

Sout

h A

frica

Peru

Para

guay

Sri L

anka

Ecua

dor

Gua

tem

ala

Indi

a

Paki

stan

Nig

eria

Gha

na

Ban

glad

esh

Cam

bodi

a

Sene

gal

Keny

a

Tanz

ania

Nep

al

Rw

anda

Uga

nda

Moz

ambi

que

Non-Internet users 192 248 806 391 536 1,278 255 484 4,150 1,575 1,177 837 1,754 1,319 788 740 840 1,316 947 1,411 855

Q What is your main reason for not using the Internet?

Arg

entin

a

Indi

a

Sene

gal

Para

guay

Rw

anda

Sout

h A

frica

Colo

mbi

a

Nig

eria

Tanz

ania

Ecua

dor

Cam

bodi

a

Moz

ambi

que

Paki

stan

Keny

a

Sri L

anka

Ban

glad

esh

Uga

nda

Peru

Gha

na

Nep

al

Gua

tem

ala

46%

70%

59% 62

% 71%

61%

52%

44%

82%

79%

66% 72

% 79%

97%

71%

61% 62

%

79% 86

%

81%

65%

19%

11%

14%

4%5%

8%13

%9%

6%15

%

3%

4%7%

2%

2%

4%

6%18

%3%

1%

1%

1% 1%1%1%

2% 3% 2%

2%

1% 3% 1%1%

1% 1%

2%

3% 2% 2%3%

1% 1%

2%

8%

6%13

%9%

20%

7%6%

9%10

%15

%4%

5%

12%

6%27

%6%

4%

6%5%

4%14

%

7%10

%9%

4%13

%6%

1%

13%

5%

3%

3%3%

2%

2%

1%

4%12

%7%

14%

15%

6%25

%4%

8%7% 7%

5%

10%

5%

26%

5%

1% 3%3%

1%

1%

1%1%

1%

2%

1%1%

1%

1% 1% 2%1%

1% 1%

1% 1% 1%

1%

1%

I don’t know what the Internet is

No interest/not useful

No access device (computer/smartphone)

I don’t know how to use it

No time, too busy

Too expensive

I am not allowed to use the Internet (restricted by family, spouse, parents etc.)Other

Main reason for not using the Internet (% of non-Internet users aged 15-65)

Figure 29.

AfterAccess

46

Base Arg

entin

a

Colo

mbi

a

Peru

Para

guay

Sri L

anka

Ecua

dor

Gua

tem

ala

Indi

a

Paki

stan

Nig

eria

Gha

na

Ban

glad

esh

Cam

bodi

a

Keny

a

Tanz

ania

Nep

al

Rw

anda

Uga

nda

Non-Internet users who own mobile phones 133 204 283 410 918 168 353 2,424 820 636 607 1,267 779 622 498 800 478 723

Q What is your main reason for not using the Internet?

Smartphone Feature phone Basic phone

Main reason for not using the Internet – “I don’t know what the Internet is”(% of non-Internet users aged 15-65 who own mobile phones)

Figure 30.

Arg

entin

a

Indi

a

Rw

anda

Colo

mbi

a

Cam

bodi

a

Ban

glad

esh

Uga

nda

Gha

na

Nep

al

Gua

tem

ala

Sri L

anka

Ecua

dor

Peru

Nig

eria

Paki

stan

Keny

a

Tanz

ania

57%

37%

32%

77%

4%18

%

62%

55%

38%

69%

66%

66%

61%

39%

34%

43%

42%44%

40%

35%

43%

81%

71%

58%

77%

65%

74%

60%

58%

32%

74%

58%

50%

77%

77%

73%

97%

97%

92%

60%

58%

33%

57%

64%

51%

79%

69%

58%

79%83

%63

%

80%

44%48

%

Para

guay

Asia Report 3.0

47

App useThe highest-reported use of mobile apps by mobile phone owners across the countries was on social networking apps such as Facebook, WhatsApp and Instagram (Table 7). For example, 70% and 71% of mobile phone owners aged 15-65 in Nepal and Cambodia, respectively, used social networking apps, compared to, 61% in Sri Lanka, 48% in India, 25% in Pakistan and just 19% in Bangladesh.

Voice and messaging apps were also popular across the board, though social media apps were the most popular in most countries.

Among the Asian countries surveyed, the most diverse use of apps (entertainment apps, business apps,

news apps etc.) was seen in India, Cambodia and Nepal. The most diverse use across the survey countries was seen in the more developed markets of Latin America, but strikingly, one third of mobile phone owners aged 15-65 in these three countries (India, Cambodia and Nepal) used news apps on mobile phones.

Pakistani and Bangladeshi Internet-enabled-mobile owners used a less diverse range of apps, sticking mainly to social media, voice and messaging apps – all at relatively low levels.

AfterAccess

48

Arg

entin

a

Colo

mbi

a

Sout

h A

frica

Peru

Para

guay

Sri L

anka

Ecua

dor

Gua

tem

ala

Indi

a

Paki

stan

Nig

eria

Gha

na

Ban

glad

esh

Cam

bodi

a

Sene

gal

Keny

a

Tanz

ania

Nep

al

Rw

anda

Uga

nda

Moz

ambi

que

Social networking apps (Facebook, WhatsApp, Instagram, Snapchat, Twitter, LinkedIn, Line etc.) 95% 94% 52% 82% 88% 61% 88% 87% 48% 25% 52% 68% 19% 71% 85% 30% 22% 70% 12% 50% 20%

Messaging or chat (text) apps (WhatsApp, Skype, Viber, Line, Talkray, Telegram, Facebook Messenger etc.) 89% 86% 45% 77% 90% 53% 86% 78% 46% 25% 43% 64% 22% 50% 85% 26% 17% 67% 16% 42% 28%

Entertainment apps (movie trailers, celebrity gossip, radio station guides etc.) 27% 43% 24% 44% 28% 29% 50% 37% 43% 13% 28% 40% 13% 51% 36% 17% 28% 38% 30% 35% 16%

Voice apps (WhatsApp, Skype, Viber, Line, Talkray etc.) 89% 86% 45% 77% 90% 57% 86% 78% 42% 24% 43% 64% 17% 27% 85% 26% 17% 41% 16% 42% 28%

Game apps (puzzles, charades etc.) 19% 32% 32% 31% 18% 28% 34% 34% 34% 15% 31% 46% 13% 37% 40% 16% 9% 34% 9% 43% 17%

News apps (local news, national headlines, technology announcements, sports etc.) 49% 41% 25% 60% 38% 31% 61% 51% 33% 12% 37% 44% 8% 36% 38% 16% 16% 37% 17% 32% 15%

Educational applications (dictionary, learning tools etc.) 34% 39% 27% 55% 25% 32% 55% 50% 30% 10% 36% 53% 8% 32% 49% 16% 12% 29% 7% 31% 14%

Search tool apps (maps, directions, phone numbers, recipes etc.) 51% 49% 30% 56% 32% 33% 59% 47% 29% 16% 28% 43% 7% 23% 39% 19% 19% 24% 18% 23% 12%

Business apps (calculate, convert, translate, etc.) 28% 22% 13% 45% 21% 21% 43% 26% 26% 14% 25% 30% 15% 51% 40% 13% 26% 11% 24% 17% 8%

Weather apps (local forecasts, natural disaster updates etc.) 57% 42% 26% 51% 46% 22% 52% 37% 22% 11% 18% 32% 3% 24% 22% 10% 4% 15% 2% 21% 12%

Trading or e-commerce apps (selling and buying online e.g.: eBay) 43% 23% 8% 36% 16% 14% 37% 18% 19% 12% 8% 18% 3% 6% 29% 9% 6% 3% 4% 10% 5%

Transport apps (public transportation info, taxis, Uber etc.) 31% 15% 10% 35% 10% 16% 33% 22% 17% 17% 6% 13% 2% 1% 20% 9% 4% 3% 5% 9% 6%

Payment gateway apps (e.g.: PayPal)*8% 15% 7% 1% 2% 4%

Base Arg

entin

a

Colo

mbi

a

Sout

h A

frica

Peru

Para

guay

Sri L

anka

Ecua

dor

Gua

tem

ala

Indi

a

Paki

stan

Nig

eria

Gha

na

Ban

glad

esh

Cam

bodi

a

Sene

gal

Keny

a

Tanz

ania

Nep

al

Rw

anda

Uga

nda

Moz

ambi

que

Feature or smartphone owners 982 1,020 1,552 972 809 849 1,039 846 1,397 571 795 458 936 878 433 1,074 789 903 660 374 667

Q Do you use these types of mobile apps on your phone?

Use of mobile apps (% of mobile phone owners aged 15-65)

Table 7.

* This question was asked only in Asian countries.

Asia Report 3.0

49

Social media

AfterAccess

50

Social media

As seen in the previous sections, social media was the most popular use of apps on mobile phones, used by as many as 70-71% of mobile phone owners in Cambodia and Nepal (Table 7). However, its use was not evenly spread across all market segments.

Gender gaps in social media use were seen across all Asian countries surveyed, with particularly large gaps in India, Bangladesh and Pakistan (Figure 31) – similar to those seen in Internet use – with Sri Lanka not too far behind.

Urban rural gaps were also seen (Figure 32) at similar levels as those in Internet use – from as high as 53% in India to as low as 23% in Pakistan.

Social media use was seen almost entirely among smartphone owners in the Asian countries surveyed (Figure 33), although some feature phone owners (e.g.: 31% and 19% aged 15-65 in Nepal and Pakistan respectively) also used social media.

The less-educated and those with lower incomes lagged behind in their use (Table 8). Social media use was most concentrated among those below the age of 35.

Social media was mostly used for keeping in touch with family and friends as well as chatting (text) and calls (Table 9). Many used it as a source of news, ranging from 52% of social media users aged 15-65 in Pakistan to 86% in Cambodia and even higher in some African countries.

The majority of social media users were willing to share their name, gender, age etc. on social media, and even contact information (Figure 34). However, users were more guarded about religious and political views as well as sexual orientation.

Bangladeshi and Indian as well as Sri Lankan social media users aged 15-65 were less trusting of the news they read on social media while users in Cambodia and Nepal were far more

trusting and Pakistani respondents were largely ambiguous (Figure 35). Approximately 40-60% of Indian, Pakistani and Bangladeshi social media users aged 15-65 stated that they did not share or forward content on social media, while a further percentage indicated that they verified the content before forwarding it (Figure 36). Nepali social media users aged 15-65 were less cautious about sharing or forwarding content on social media.

Asia Report 3.0

51

19%

15%

Gender gaps in social media use (% of population aged 15-65)

Base Arg

entin

a

Colo

mbi

a

Sout

h A

frica

Peru

Para

guay

Sri L

anka

Ecua

dor

Gua

tem

ala

Indi

a

Paki

stan

Nig

eria

Gha

na

Mya

nmar

Ban

glad

esh

Cam

bodi

a

Leso

tho

Sene

gal

Keny

a

Tanz

ania

Nep

al

Rw

anda

Uga

nda

Moz

ambi

que

All respondents(Male) 478 487 795 508 879 834 511 656 2,478 1,060 912 547 3,818 1,092 735 515 603 544 531 912 556 848 527

All respondents(Female) 762 938 1,020 970 478 1,183 909 751 2,591 942 896 653 3,386 928 1,388 1,329 578 664 669 1,096 655 909 644

Q Do you use social media like Facebook, WhatsApp, Twitter etc.?

GapFemaleMale

Gender gap in social media use %

Male social media users(% of male population)= ÷( )- Female social media users

(% of female population)Male social media users(% of male population)

Sri L

anka

Ecua

dor

Nig

eria

Paki

stan

Keny

a

Tanz

ania

Sene

gal

Arg

entin

a

Indi

a

Mya

nmar

Rw

anda

Sout

h A

frica

Cam

bodi

a

Moz

ambi

que

Colo

mbi

a

Ban

glad

esh

Uga

nda

Peru

Para

guay

Gha

na

Nep

al

Gua

tem

ala

Leso

tho

Figure 31.80

%

86%

46%

74%

61%

39%

86%

71%

22%

18%

37%

35% 37

%

35%

39%

31%

16%

38%

8%

46%

13%

83%

77%

43%

64%

62%

21%

77%

55%

9% 9%

20% 23

%

18%

6%

25%

31%

25%

22%

11%

27%

5%

41%

6%

-4%

10%6%

14%

-2%

46%

11%

22%

60%50% 47%

33%

66%

22%32%

11%

36%29% 30% 30%

45%

10%

51%

AfterAccess

52

Urban-rural gaps in social media use (% of population aged 15-65)

Base Arg

entin

a

Colo

mbi

a

Sout

h A

frica

Peru

Para

guay

Sri L

anka

Ecua

dor

Gua

tem

ala

Indi

a

Paki

stan

Nig

eria

Gha

na

Mya

nmar

Ban

glad

esh

Cam

bodi

a

Leso

tho

Sene

gal

Keny

a

Tanz

ania

Nep

al

Rw

anda

Uga

nda

Moz

ambi

que

All respondents(Urban) 1,208 986 1,050 1,178 824 803 932 550 2,200 793 1,147 721 3,477 808 897 844 790 727 720 940 711 1,024 718

All respondents(Rural) 32 439 765 300 533 1,214 488 857 2,869 1,209 661 479 3,727 1,212 1,226 1,000 391 481 480 538 500 733 453

Q Do you use social media like Facebook, WhatsApp, Twitter etc.?

GapRuralUrban

Urban-rural gap in social media use %

Urban social media users(% of urban population)= ÷( )- Rural social media users

(% of rural population)Urban social media users(% of urban population)

Figure 32.

26%

12%

Sri L

anka

Ecua

dor

Nig

eria

Paki

stan

Keny

a

Tanz

ania

Sene

gal

Arg

entin

a*

Indi

a

Mya

nmar

Rw

anda

Sout

h A

frica

Cam

bodi

a

Moz

ambi

que

Colo

mbi

a

Ban

glad

esh

Uga

nda

Peru

Para

guay

Gha

na

Nep

al

Gua

tem

ala

Leso

tho

81%

88%

52%

74%

73%

37%

82%

67%

24%

16%

41%

37%

46% 48

%

40%

49%

30%

36%

20%

30%

21%

92%

79%

31%

46%

43%

27%

77%

58%

11% 12

%

20%

17%

18%

11%

25%

17%

23%

17%

5%

24%

2%

8%

3%

-13%

10%

40% 38% 41%

27%

5%

14%

53%

23%

52% 54% 52%46%

64%

42%

66%

84%

33%

88%

72%

84%

40%

* Low rural base

Asia Report 3.0

53

Base Arg

entin

a

Colo

mbi

a

Sout

h A

frica

Peru

Para

guay

Sri L

anka

Ecua

dor

Gua

tem

ala

Indi

a

Paki

stan

Nig

eria

Gha

na

Ban

glad

esh

Cam

bodi

a

Sene

gal

Keny

a

Tanz

ania

Nep

al

Rw

anda

Uga

nda

Moz

ambi

que

Mobile phone owners 1,116 1,297 1,398 1,234 1,209 1,609 1,191 1,214 3,252 1,208 1,123 901 1,531 1,526 969 1,054 761 1,478 635 1,031 632

Q Do you use social media like Facebook, WhatsApp, Twitter etc.?

Arg

entin

a

Colo

mbi

a

Sout

h A

frica

Peru

Para

guay

Sri L

anka

Ecua

dor

Gua

tem

ala

Indi

a

Gha

na

Ban

glad

esh

Cam

bodi

a

Sene

gal

Keny

a

Tanz

ania

Nep

al

Rw

anda

Uga

nda

Moz

ambi

que

Social media use (% of mobile phone owners aged 15-65) Figure 33.

4%47

%85

%

23%

76%

97%

59%

90%

89%

8%45

%75

%

21%

61%

95%

3%69

%98

%

4%4%72

%

28%

59%

97%

16%

85%

91%

3%6%

72%

8%19

%62

%

5%31

%86

%

2%4%

62%

6%3%

82%

20%

34%

54%

2%30

%85

%

2%15

%83

%

2%31

%77

%

3%8%

85%

37%

50%

67%

Smartphone Feature phone Basic phone

Paki

stan

Nig

eria

14%

38%

88%

AfterAccess

54

Arg

entin

a

Peru

Para

guay

Sri L

anka

Ecua

dor

Gua

tem

ala

Indi

a

Paki

stan

Nig

eria

Gha

na

Ban

glad

esh

Cam

bodi

a

Leso

tho

Sene

gal

Keny

a

Tanz

ania

Nep

al

Rw

anda

Educ

atio

n Secondary education or higher 82% 73% 88% 36% 88% 81% 29% 42% 51% 52% 22% 64% 74% 64% 43% 26% 59% 24%

Primary or none 15% 34% 40% 9% 69% 40% 5% 11% 3% 9% 6% 19% 28% 14% 5% 3% 9% 1%

Inco

me

Above average 83% 75% 75% 41% 86% 64% 22% 26% 37% 39% 15% 40% 49% 38% 43% 31% 44% 20%

Below average 79% 58% 49% 19% 80% 57% 10% 10% 25% 27% 11% 25% 26% 29% 21% 7% 35% 4%

Zero income 86% 77% 65% 22% 88% 69% 15% 6% 0% 0% 11% 13% 23% 0% 0% 1% 28% 5%

Age

15-25 91% 90% 89% 52% 93% 85% 29% 19% 32% 38% 22% 48% 51% 45% 40% 14% 47% 6%

26-35 90% 79% 75% 44% 87% 66% 15% 13% 34% 36% 14% 41% 38% 43% 27% 23% 40% 12%

36-45 85% 55% 63% 28% 78% 49% 10% 9% 25% 18% 9% 23% 21% 25% 16% 8% 25% 7%

46-55 73% 43% 44% 12% 56% 33% 6% 7% 17% 16% 3% 11% 7% 9% 9% 7% 10% 1%

56-65 52% 28% 22% 6% 36% 22% 2% 11% 9% 7% 1% 8% 6% 6% 7% 3% 3% 1%

Base Arg

entin

a

Peru

Para

guay

Sri L

anka

Ecua

dor

Gua

tem

ala

Indi

a

Paki

stan

Nig

eria

Gha

na

Ban

glad

esh

Cam

bodi

a

Leso

tho

Sene

gal

Keny

a

Tanz

ania

Nep

al

Rw

anda

All respondents 1,240 1,478 1,357 2,017 1,420 1,407 5,069 2,002 1,706 1,145 2,020 2,123 1,844 1,181 1,179 1,102 2,008 1,118

Q Do you use social media like Facebook, WhatsApp, Twitter etc.?

Social media use, by education level, income level and by age group (% of population aged 15-65)

Table 8.

Asia Report 3.0

55

Sout

h A

frica

Sri L

anka

Indi

a

Paki

stan

Nig

eria

Gha

na

Ban

glad

esh

Cam

bodi

a

Sene

gal

Keny

a

Tanz

ania

Nep

al

Rw

anda

Uga

nda

Moz

ambi

que

Chatting (text) 93% 76% 86% 73% 97% 95% 93% 72% 81% 96% 99% 97% 98% 49% 93%

Staying in contact with friends and family 77% 76% 91% 72% 96% 92% 94% 79% 96% 92% 86% 93% 90% 46% 90%

Making calls 73% 80% 83% 70% 58% 81% 78% 76% 94% 62% 72% 90% 79% 24% 73%

Sharing videos/pictures/music 65% 68% 74% 63% 83% 86% 67% 67% 90% 84% 88% 87% 82% 27% 80%

Making new friends 63% 66% 68% 58% 90% 93% 76% 61% 91% 91% 90% 80% 85% 24% 83%

Reading news62% 78% 77% 52% 86% 80% 73% 86% 89% 79% 94% 81% 89% 21% 74%

Playing games45% 49% 66% 54% 41% 53% 42% 32% 55% 46% 38% 42% 45% 13% 47%

Looking for educational content44% 63% 71% 51% 74% 76% 57% 50% 71% 71% 57% 42% 49% 9% 71%

Getting opinions/sharing experiences43% 62% 63% 48% 76% 73% 57% 54% 56% 80% 70% 61% 67% 15% 52%

Making professional and business contacts29% 37% 57% 38% 47% 46% 37% 26% 65% 51% 42% 14% 43% 5% 37%

Following government social media pages (to look for jobs or updates on policies) 25% 38% 58% 39% 46% 44% 35% 40% 54% 55% 54% 27% 54% 2% 24%

Following local politicians21% 38% 47% 41% 31% 34% 33% 43% 49% 56% 54% 29% 54% 4% 18%

Sharing own produced content21% 49% 55% 45% 27% 43% 52% 50% 48% 41% 36% 9% 43% 2% 46%

Marketing products/services13% 18% 45% 34% 25% 27% 29% 9% 46% 31% 21% 8% 18% 2% 20%

Base Sout

h A

frica

Sri L

anka

Indi

a

Paki

stan

Nig

eria

Gha

na

Ban

glad

esh

Cam

bodi

a

Sene

gal

Keny

a

Tanz

ania

Nep

al

Rw

anda

Uga

nda

Moz

ambi

que

Social media users 669 593 754 369 496 323 251 680 387 423 241 648 156 312 230

Q What do you use social media for?

What social media is used for (% of social media users aged 15-65)

Table 9.

AfterAccess

56

95%

90%

87%

80%

80%

36%

31%

23%

21%

Argentina

87% 89

%83

%76

%80

%37

%44

%17

% 20%

Colombia

63%

74%

57%

47%

54%

37%

28%

11%

11%

South Africa

95%

89%

86%

80%

85%

41%

49%

30%

25%

Peru

93%

93%

88%

81%

86%

45%

29%

16%

14%

Paraguay

90%

88%

88%

81%

77% 78

%70

%33

%27

%

Sri Lanka

92%

85%

80%

76%

83%

42% 45

%24

%23

%

Ecuador

93%

82%

80%

70% 73

%41

%52

%20

%16

%

Guatemala83

% 85%

80%

71%

63%

71%

57%

40%

26%

India

61%

67%

65%

56%

52% 55

%62

%48

%32

%

Pakistan

99%

90%

76%

85% 87

%80

%79

%43

%22

%

Nigeria

Types of information shared on social media (% of social media users aged 15-65)

Figure 34.

Gender

Real name

Age

Marital status

Pictures or videos of you and your family and friends

Mobile number/email address

Religion

Political views

Sexual orientation

Asia Report 3.0

57

91%

83%

71%

59%

87%

67% 71

%22

%23

%

Ghana

96%

96%

92%

77%

76%

69%

78%

33%

23%

Bangladesh

75%

68%

52%

45%

84%

66%

29%

11%

10%

Cambodia

93%

94%

69%

69%

75%

60%

54%

24%

10%

Senegal

90%

80%

67%

72%

78%

62%

69%

41%

22%

Kenya

88%

84%

56%

64%

79%

45%

45%

19%

5%

Tanzania

96%

93%

85%

84%

90%

43%

51%

28%

8%

Nepal

90%

90%

78%

83% 84

%53

%66

%24

% 27%

Rwanda78

%84

%69

%53

%60

%46

%34

%9%

6%

Uganda

74%

69% 73

%63

%70

%41

%32

%10

%8%

Mozambique

Base Arg

entin

a

Colo

mbi

a

Sout

h A

frica

Peru

Para

guay

Sri L

anka

Ecua

dor

Gua

tem

ala

Indi

a

Paki

stan

Nig

eria

Gha

na

Ban

glad

esh

Cam

bodi

a

Sene

gal

Keny

a

Tanz

ania

Nep

al

Rw

anda

Uga

nda

Moz

ambi

que

Social media users 993 1,246 669 999 802 593 1,122 878 754 369 496 323 251 680 387 423 241 648 156 312 230

Q What information do you share on social media ?

AfterAccess

58

19%

30%

28%

19%

4%

Sout

h A

frica

25%

40%

14%

19%

2%

Sri L

anka

32%

26%

13%

25%

4%

Indi

a

11%

18%

50%

20%

1%

Paki

stan

19%

42%

14%

21%

4%

Nig

eria

13%

31%

24%

30%

2%

Gha

na

53%

13%

8%23

%

3%

Ban

glad

esh

7%33

%14

%43

%

3%

Cam

bodi

a

26%

38%

24%

11%

1%

Sene

gal

22%

34%

16%

27%

1%

Keny

a

9%41

%11

%38

%

1%

Tanz

ania

11%

17%

14%

58%

1%

Nep

al

7%30

%19

%41

%

3%

Uga

nda

20%

29%

39%

12%

Moz

ambi

que

Strongly do not trust Do not trust TrustDon't know Strongly trust

Degree of trust in news read on social media (% of social media users aged 15-65)

Figure 35.

Base Sout

h A

frica

Sri L

anka

Indi

a

Paki

stan

Nig

eria

Gha

na

Ban

glad

esh

Cam

bodi

a

Sene

gal

Keny

a

Tanz

ania

Nep

al

Uga

nda

Moz

ambi

que

Social media users 669 593 754 369 496 323 251 680 387 423 241 648 312 230

Q Can you trust news you read on social media (Facebook etc.)?

Asia Report 3.0

59

I don’t share or forward messages

I always verify its truthfulness before sharing

I share it if it is from a trusted friend or source

I share it without checking

Sharing content on social media (% of social media users aged 15-65)

Figure 36.50

%6%

12%

32%

31%

9%21

%39

%

17%

51%

22%

10%

45%

6%9%

40%

50%

7%7%

36%

59%

27%

6%8%

37%

10%

13%

41%

41%

40%

13%

5%

61%

15%

17%

7%

16%

61%

18%

5%

58%

32%

6%5%

27%

52%

15%

6%

19%

58%

22%

25%

45%

24%

6%

25%

49%

22%

4%

27%

38%

31%

4%

11%

65%

20%

4%

39%

17%

42%

2%

10%

72%

16%

Arg

entin

a

Colo

mbi

a

Sout

h A

frica

Peru

Para

guay

Sri L

anka

Ecua

dor

Gua

tem

ala

Indi

a

Paki

stan

Nig

eria

Gha

na

Ban

glad

esh

Cam

bodi

a

Sene

gal

Keny

a

Nep

al

Uga

nda

Moz

ambi

que

2%

Base Arg

entin

a

Colo

mbi

a

Sout

h A

frica

Peru

Para

guay

Sri L

anka

Ecua

dor

Gua

tem

ala

Indi

a

Paki

stan

Nig

eria

Gha

na

Ban

glad

esh

Cam

bodi

a

Sene

gal

Keny

a

Nep

al

Uga

nda

Moz

ambi

que

Social media users 993 1,246 669 999 802 593 1,122 878 754 369 496 323 251 680 387 423 648 312 230

Q When you share news that is on your newsfeed or forward messages that you receive, do you:

AfterAccess

60

Asia Report 3.0

61

Public Wi-Fi

AfterAccess

62

Public Wi-FiPublic Wi-Fi was used by 78% of Internet users aged 15-65 in Pakistan, 53% in Nepal, 42% in India, 40% in Cambodia, 22% in Sri Lanka and 21% in Bangladesh (Figure 37). Free Wi-Fi was utilized by at least half of all these respondents, but was most popular among those in Pakistan. Users of other forms of public Wi-Fi used paid and/or conditional hotspots.

On the whole, a larger share of rural Internet users made use of the free public Wi-Fi than urban Internet users (Figure 38). The latter more often used paid hotspots. More male Internet users made use of public Wi-Fi than female (Figure 39) and more respondents from higher SECs than lower (Figure 40). The relationship between age and Wi-Fi use appears to be somewhat U-shaped (Figure 41). The youngest (15-25) and oldest (55-65) use free public Wi-Fi more often than the middle age groups (i.e.: those earning, with less time and flexibility, etc.). Sri Lanka India Pakistan Bangladesh Cambodia

11%

7% 1%1%

1%

78%

23%

13% 3%

3%1%

58%

39%

25%

4%

6%

4%

22%

8%8%

2%2%

79%

14%

3%

17%

2%3%

60%

12%

16%

8%10

%7%

47%

Nepal

Public Wi-Fi use (% of Internet users aged 15-65)

Figure 37.

Base Sri L

anka

Indi

a

Paki

stan

Ban

glad

esh

Cam

bodi

a

Nep

alInternet users 739 919 427 266 804 692

Q Do you access the Internet over public Wi-Fi through the following means?

Conditional Wi-Fi (shops, cafes etc.)

Free public Wi-Fi

No, I don’t use public Wi-Fi

Other

Combination of the above

Paid Wi-Fi (hotspots)

Asia Report 3.0

63

Public Wi-Fi use, by location (% of Internet users aged 15-65)

Figure 38.

Base

Sri Lanka India Pakistan Bangladesh Cambodia Nepal

Urban Rural Urban Rural Urban Rural Urban Rural Urban Rural Urban Rural

Internet users 348 391 476 443 211 216 146 120 456 348 478 214

Base

Sri Lanka India Pakistan Bangladesh Cambodia Nepal

Urban Rural Urban Rural Urban Rural Urban Rural Urban Rural Urban Rural

Internet users 348 391 476 443 211 216 146 120 456 348 478 214

Q Do you access the Internet over public Wi-Fi through the following means?

16%

7%7%

10%

7%53

%

Rural

Nepal

11%

19%

9%10

%7%

45%

Urban

Cambodia

20%

2%

23%

3%3%

49%

Urban

11%

14%

2%3%

66%

Rural

Bangladesh

10%

10%

5%

1%

75%

Urban

7%6%

1%3%

82%

Rural

Pakistan

35%

38%

1%2%

4%

21%

Urban

42%

16%

5%9%

4%

23%

Rural

India

19%

19%

3%4%

1%

55%

Urban

27%

7% 3%2%

1%

60%

Rural

Sri Lanka

2%

11%

7%79

%

Urban

12%

6% 1%2%

1%

78%

Rural

4%

Conditional Wi-Fi (shops, cafes etc.)

Free public Wi-Fi

No, I don’t use public Wi-Fi

Other

Combination of the above

Paid Wi-Fi (hotspots)

AfterAccess

64

Public Wi-Fi use, by gender (% of Internet users aged 15-65)

Figure 39.

Base

Sri Lanka India Pakistan Bangladesh Cambodia

Male Female Male Female Male Female Male Female Male Female

Internet users 390 349 652 267 283 144 199 67 365 439

Q Do you access the Internet over public Wi-Fi through the following means?

* Low base for Bangladesh women

Sri Lanka

14%

5%

2%

79%

Male

9%8% 3%

1%2%

78%

Female

India

24%

12%

3%4%

56%

Male

19%

14% 3%

1%2%

60%

Female

Pakistan

38%

26%

2%

7%5%

22%

Male

42%

22%

7%5%

1%

22%

Female

Bangladesh*

10%

9%

3%

78%

Male3%

4%9%84

%

Female

Cambodia

13%

4%

28%

2%2%

51%

Male

15%

3%

8%

3%3%

68%

Female

Conditional Wi-Fi (shops, cafes etc.)

Free public Wi-Fi

No, I don’t use public Wi-Fi

Other

Combination of the above

Paid Wi-Fi (hotspots)

Asia Report 3.0

65

Public Wi-Fi use, by SEC (% of Internet users aged 15-65)

Figure 40.

* Low base for Bangladesh SEC D-E

Base

Sri Lanka India Pakistan Bangladesh Cambodia Nepal

A-C D-E A-C D-E A-C D-E A-C D-E A-C D-E A-C D-E

Internet users 572 167 809 109 207 220 217 49 196 595 418 274

Q Do you access the Internet over public Wi-Fi through the following means?

5%7%

2%2%

1%

82%

Sri Lanka

13%

6%

1%1%

1%

77%

A-C D-EIndia

23%

15% 3%

3%1%

56%

A-C

19%

3%

5%

3%1%

69%

D-E

Pakistan

23%

35%

8%11

%

2%

21%

A-C

48%

20%

1%4%

4%

23%

D-E D-E

10%

7%5%

79%

Bangladesh*

8%8%

3%1%

80%

A-C

Cambodia

13%

4%

28%

2%2%

51%

A-C

15%

3%

8%

3%3%

68%

D-E

Nepal

13%

20%

9%11

%

4%

43%

A-C

11%

11%

6%8%

11%

53%

D-E

Conditional Wi-Fi (shops, cafes etc.)

Free public Wi-Fi

No, I don’t use public Wi-Fi

Other

Combination of the above

Paid Wi-Fi (hotspots)

AfterAccess

66

Public Wi-Fi use, by age group (% of Internet users aged 15-65) Figure 41.

Base Sri Lanka India Pakistan Bangladesh Cambodia Nepal

Internet users (15-25) 188 441 191 111 309 276

Internet users (26-35) 174 261 116 84 283 264

Internet users (36-45)) 142 135 74 51 127 116

Internet users (46-55)) 55 60 28 14 54 30

Internet users (56-65) 28 22 18 6 31 6

Q Do you access the Internet over public Wi-Fi through the following means?

1%2%

1%14

%7%

74%

15 -

25

4%2%

1%1%

10%

81%

26 -

35

10%

8%81

%36

- 45

10%

86%

3%1%

46 -

55

18%

8%71

%2%

2%56

- 65

Sri Lanka

24%

11%

60%

2%3%

1%15

- 25

22%

14%

7%52

%3%

1%26

- 35

17%

18%

60%

4%36

- 45

22%

14%

8%51

%5%

46 -

55

27%

15%

54%

3%56

- 65

India

37%

20%

7%29

%2%

3%15

- 25

56%

28%

9%2%

4%2%

26 -

35

30%

33%

12%

19%

3%4%

36 -

45

2%31

%31

%35

%1%

46 -

55

36%

38%

23%

3%56

- 65

Pakistan

11%

6%77

%2%

3%15

- 25

12%

77%

5%4%

2%26

- 35

10%

86%

3%1%

36 -

45

94%

3%3%

46 -

55

7%93

%56

- 65

Bangladesh

17%

18%

55%

4%3%

2%15

- 25

11%

21%

57%

5%3%

3%26

- 35

10%

14%

70%

1%4%

36 -

45

11%

6%78

%5%

46 -

55

11%

8%81

%1%

56 -

65

Cambodia

12%

18%

8%11

%8%

43%

15 -

25

13%

15%

9%6%

7%51

%26

- 35

3%11

%15

%6%

17%

47%

36 -

45

2%20

%9%

13%

6%50

%46

- 55

14%

19%

66%

56 -

65

Nepal

Combination of the above

Conditional Wi-Fi (shops, cafes…)

Paid Wi-Fi (hotspots)

Free Public Wi-Fi

No, I don’t use public Wi-Fi

Other (specify)

Note: Bases are low for some age groups (indicated in red)

Asia Report 3.0

67

Mobile phoneexpenditure

AfterAccess

68

Mobile phone expenditureThe average amount spent monthly on services (including voice, SMS and data) by mobile phone owners in the Asian countries surveyed was lower than that spent by users from the Latin American and African countries surveyed (Figure 42). Indian mobile phone owners spent USD2.3 on average at the time of survey (October-November 2017) – the lowest from the survey countries. Bangladeshi owners spent the second lowest (USD 2.9) from among the Asian countries, and Nepali the third lowest (USD 3.2).

It should be noted that the expenditure amounts captured through a demand-side survey (i.e.:

from the user, rather than from the supply-side or the network provider) might differ largely from industry ARPU numbers. For example, at the end of September 2017 the Indian ARPU was INR896, or USD 1.37. The discrepancy can be accounted for by the fact that in the demand side data, the average is calculated based on the number of unique subscribers, while APRU (supply-side) is calculated based on the number of active SIMs, which we know from the survey is a larger number than that of unique subscribers (e.g.: 26% of mobile phone owners in India had more than one SIM). Another difference is that the survey target group consists of

those aged 15-65 only, while ARPU calculations make no such distinction.

Table 10 provides expenditure data as captured by the survey, disaggregated by urban-rural, gender, income and age groups as well as handset types.

6 https://coai.com/sites/default/files/COAI%20ARPU%20REPORT%20Q2%20FY%202017-18%20Final.pdf

Asia Report 3.0

69

17.5

Arg

entin

a

10.2

Sout

h A

frica

12.0

Peru

13.2

Para

guay

5.5

Sri L

anka

13.5

Gua

tem

ala

2.3

Indi

a

3.3

Paki

stan

7.1

Nig

eria

6.6

Gha

na

2.9

Ban

glad

esh

4.1

Cam

bodi

a

9.3

Keny

a

4.6

Tanz

ania

3.2

Nep

al

2.5

Rw

anda

3.1

Moz

ambi

que

Average monthly expenditure on mobile services (USD)Figure 42.

Q Could you tell me how much you spent last month for voice, SMS and data in total (airtime and subscription)?

Base Arg

entin

a

Sout

h A

frica

Peru

Para

guay

Sri L

anka

Gua

tem

ala

Indi

a

Paki

stan

Nig

eria

Gha

na

Ban

glad

esh

Cam

bodi

a

Keny

a

Tanz

ania

Nep

al

Rw

anda

Moz

ambi

que

Mobile phone owners 1,116 1,398 1,234 1,209 1,609 1,214 3,252 1,208 1,123 901 1,531 1,526 1,054 761 1,478 635 632

AfterAccess

70

Arg

entin

a

Sout

h A

frica

Peru

Para

guay

Sri L

anka

Gua

tem

ala

Indi

a

Paki

stan

Nig

eria

Gha

na

Ban

glad

esh

Cam

bodi

a

Keny

a

Tanz

ania

Nep

al

Rw

anda

Moz

ambi

que

Loca

tion Urban 17.4 18.7 12.9 15.3 8.8 13.7 4.4 2.7 8.5 11.9 4.1 7 17.7 7.3 7.6 6.8 5.8

Rural 23.3 7.7 8.7 9.3 6.7 13.4 2.4 5.1 8 6.3 3 5 9 4 6.3 2.1 2.2

Gen

der Male 19.2 18.3 14.8 15.5 8.7 16.5 3.7 4.4 9.7 8.3 3.9 6 13.6 6.1 7.6 4.2 4.1

Female 15.9 12.5 10.3 12 5.6 10.9 2.4 3.9 6.5 11.4 2.3 5.3 9.6 4.6 6.8 2.4 3.5

Inco

me

leve

l Above average 22.2 35.8 17.4 17.5 9.9 18.6 4.7 7.8 13.6 27 4.1 7.2 27 8.5 10.4 8.3 6.1

Below average 14.9 7.9 8.3 10 4.5 9.2 1.7 1.8 4.7 6.3 2.2 4.6 7.2 3.1 6.9 1.9 1.5

Zero income 11.4 5.1 8.3 10.6 4.2 11 2.6 2.5 3.1 0.5 6.5 1.3 1.6

Age

grou

p (y

ears

)

15-25 15 9.8 10.9 14.5 8.2 13.1 3.8 6 5.3 9.4 3.4 6.2 10 5.4 7.1 3.7 3.3

26-35 17.6 14.5 13 15.2 7.8 15.6 3.3 4.7 9.7 9.2 3.5 6.7 10 6.3 7.8 3.5 4.1

36-45 21.4 23.8 12.3 13.3 7.8 14 2.7 2.7 10 11.7 3.3 5.7 13.5 4.2 6.6 4.5 3.6

46-55 17.1 18.2 12.1 11 5.7 12.1 3.3 2.2 10.1 7.9 3 4.5 12.1 5 7.8 2.1 3.7

56-65 17.8 11.7 11.5 10.9 5.3 12.4 2.2 2.7 8 11.1 2.6 3.4 18.2 5.7 7.4 1.8 7.8

Type

of h

ands

et Basic phone 11.3 5 6.5 7.5 3.7 8.4 2.1 3.8 5.4 6.8 2.4 3.5 7.3 4 6 2.1 2.3

Feature phone 14 9.4 9.6 13.3 3.9 14 2.1 4.6 8.9 7.4 2.5 3.9 9.1 3.1 9.5 2.2 6.3

Smartphone 18.9 21.8 14.6 15.9 10.7 16.2 6 5 10.8 15.2 6 7.8 21.2 10.7 8.2 16 9.5

Q Could you tell me how much you spent last month for voice, SMS and data in total (airtime and subscription)?

Average monthly expenditure on mobile services, by location, gender, income level, age group and type of handset (USD)

Table 10.

Base Arg

entin

a

Sout

h A

frica

Peru

Para

guay

Sri L

anka

Gua

tem

ala

Indi

a

Paki

stan

Nig

eria

Gha

na

Ban

glad

esh

Cam

bodi

a

Keny

a

Tanz

ania

Nep

al

Rw

anda

Moz

ambi

que

Mobile phone owners 1,116 1,398 1,234 1,209 1,609 1,214 3,252 1,208 1,123 901 1,531 1,526 1,054 761 1,478 635 632

Asia Report 3.0

71

Online harrassment

AfterAccess

72

Online harassmentOf Internet users aged 15-65, over a quarter in Cambodia, 19% in India, 12% in Pakistan and Bangladesh, 9% in Sri Lanka and 4% in Nepal indicated that they had experienced online harassment before (Figure 43). In India and Bangladesh, experiences with online harassment were higher among rural Internet users, while in Cambodia, urban Internet users faced higher levels of harassment (Figure 44). The other Asian survey countries did not show a significant urban-rural gap.

Gender-wise disaggregation of the data showed that, except for Cambodia, more men reported having experienced online harassment than women (Figure 45). In Cambodia, the pattern was reversed with up to 29% of female Internet users aged 15-65 having experienced harassment as compared to 23% of their male counterparts.

The most common form of online harassment experienced by Indian victims was being called offensive names, while in Cambodia it was being cyber stalked (Figure 46)7. In both countries, the most common platform on which harassment was experienced was social media, though Indian respondents also mentioned chat applications and website comments sections (Figure 47).

Half of the Internet users in Cambodia who experienced harassment indicated that they did not know the perpetrator – either offline or online – while a quarter each said they either knew the person offline or they only knew them online (Figure 48). In India, a third each of respondents indicated that the perpetrator was from one of these three groups of people (unknown offline or online,

known online but not offline, known both online and offline). When asked what effect the harassment had on the respondent, the most common response was “It had no effect” (Figure 49). However, 28% of Indian respondents who experienced online harassment said the incident/s reduced their use of the particular website.

7 Bases for the other four Asian countries become too low to analyze harassment from here on.

Asia Report 3.0

73

Experienced online harassment before (% of Internet users aged 15-65 )

Figure 43.

Base Sri L

anka

Indi

a

Paki

stan

Ban

glad

esh

Cam

bodi

a

Nep

al

Internet users 763 919 427 266 804 765

Q1 Being called offensive namesQ2 Being purposefully embarrassed or criticized in another way (besides being called offensive names)Q3 Being physically threatenedQ4 Being sexually harassedQ5 Being approached repeatedly by unwanted contacts (cyber-stalked)

9%

Sri Lanka

19%

India

12%

Bangladesh

26%

Cambodia

4%

Nepal

12%

Pakistan

AfterAccess

74

Experienced online harassment, by location (% of Internet users aged 15-65 )

Experienced online harassment, by gender(% of Internet users aged 15-65 )

Figure 44. Figure 45.

Ban

glad

esh*

Nep

al

Indi

a

Sri L

anka

Paki

stan

Cam

bodi

a

7%

10%

20%

17%

13%

9%

13%

10%

23%

29%

4% 3%

Ban

glad

esh

Nep

al

Indi

a

Sri L

anka

Paki

stan

Cam

bodi

a

7%

9%

17%

20%

12%

12%

9%

14%

28%

25%

4%

3%

Base Sri L

anka

Indi

a

Paki

stan

Ban

glad

esh

Cam

bodi

a

Nep

al

Internet users (Male) 390 652 283 199 365 384

Internet users (Female) 349 267 144 67 439 308

Base Sri L

anka

Indi

a

Paki

stan

Ban

glad

esh

Cam

bodi

a

Nep

al

Internet users (Urban) 348 476 211 146 456 478

Internet users (Rural) 391 443 216 120 348 214

Urban Rural Male Female

Q1 Being called offensive namesQ2 Being purposefully embarrassed or criticized in another way (besides being called offensive names)Q3 Being physically threatenedQ4 Being sexually harassedQ5 Being approached repeatedly by unwanted contacts (cyber-stalked)

* Low base for Bangladesh women

Asia Report 3.0

75

Base Indi

a

Cam

bodi

a

Respondents who faced online harassment 178 197

Base Indi

a

Cam

bodi

a

Respondents who faced online harassment 178 197

Q Which form of harassment did you most recently experience, personally?

Q On which type of platform did you experience this harassment?

Forms of harassment experienced (% of respondents who experienced online harassment)

Platform on which the harassment was experienced (% of respondents who experienced online harassment)

Figure 46.

Figure 47.

India Cambodia

India Cambodia

39%

29%

16%

7%

3%

6%

85%

4%

0%

0% 1%

10%

14% 17

%

2% 1%

67%

49%

20%

13%

3%

16%

Being physically threatened

Comments section of a website

Being sexually harassed

Online gaming platforms

Being approached repeatedly by unwanted contacts (cyber-stalked)

Other

Email

Being called offensive names

Social media (e.g.: Facebook)

Being purposefully embarassed or criticized in any other way(besides being called offensive names)

Chat application (e.g.: WhatsApp)

AfterAccess

76

Source of harassment (% of Internet users who have experienced online harassment)

Effect of harassment on Internet use (% of Internet users who have experienced online harassment)

Figure 48.

Figure 49.

Someone I've met offline before

One of my online contacts/ friends that I've never met offline

Someone I don't know at all (online or offline)

33%

27%

33%

49%

33%

24%

India Cambodia

India Cambodia

48%

62%

28%

11%

15%

7%5%

11%

4%

9%

It has had no effect

I reduced use of the particular website

I deleted the app or my profile

I unfriended/ blocked contacts or left that group/ forum

I now limit my use of the Internet as a whole

Base Indi

a

Cam

bodi

a

Respondents who faced online harassment 178 197

Base Indi

a

Cam

bodi

a

Respondents who faced online harassment 178 197

Q What effect has this had on your use of the Internet?

Q Who was the source of this harassment?

Asia Report 3.0

77

E-commerce

AfterAccess

78

Mobile moneyOn a percentage basis, the use of mobile phones for financial transactions (i.e.: to send or receive money) was very low in the Asian survey countries. For example, in India, only 6% of mobile phone owners aged 15-65 said yes when asked if they ever used their mobile phone for financial transactions. Bangladesh had the highest use of 30% among the Asian survey countries (Figure 50).

It is noted that in some cases, the data presented in the graphs, based on the survey questions, does not tally with supply-side data, for example Indian supply side data indicates that at the time of survey, Paytm users alone should have accounted for 15% of the Indian population, whereas according to survey responses, mobile money is used by 5% of mobile phone owners aged 15-65. The capping of the age group which this survey captures could lead to some amount of under-estimation.

Mobile money use (% of mobile phone owners aged 15-65)

Figure 50.

Base Arg

entin

a

Colo

mbi

a

Sout

h A

frica

Peru

Para

guay

Sri L

anka

Ecua

dor

Gua

tem

ala

Indi

a

Paki

stan

Nig

eria

Gha

na

Ban

glad

esh

Cam

bodi

a

Leso

tho

Sene

gal

Keny

a

Nep

al

Uga

nda

Moz

ambi

que

Mobile phone owners 1,116 1,297 1,398 1,234 1,209 1,609 1,191 1,214 3,252 1,208 1,123 901 1,531 1,526 1,708 969 1,054 1,478 1,031 632

Q Do you ever use your mobile phone for financial transactions: to send or receive money?

Arg

entin

a

Colo

mbi

a

Sout

h A

frica

Peru

Para

guay

Sri L

anka

Ecua

dor

Gua

tem

ala

Indi

a

Paki

stan

Nig

eria

Gha

na

Ban

glad

esh

Cam

bodi

a

Leso

tho

Sene

gal

Keny

a

Nep

al

Uga

nda

Moz

ambi

que

7%

12%

19%

5%

54%

5%

9%

6%6%

13%

12%

38%

30%

51%

69%

5%

86%

80%

4%

33%

Asia Report 3.0

79

PlatformsUsing platforms for buying goods and servicesInternet users were asked if they were aware of Internet websites or mobile apps to buy and sell goods or services that they need – essentially digital platforms. This could include apps such as Uber, Lyft, AliExpress and Upwork.com, as well as social media such as Facebook and Instagram.

The responses showed that Sri Lankan and Indian Internet users had the highest levels of awareness among respondents from the Asian countries, with Cambodian respondents coming in third (Figure 51). Awareness was highest with regard to buying and selling goods and services (through platforms such as Amazon, AliExpress, eBay, etc.), buying/selling tickets and appointments, and transport and ridesharing apps. Almost half of Sri Lanka’s and India’s Internet users aged 15-65 were aware of these. When it came to actual use (Figure 52) as well, India had the highest number of users (out of those aware of the platforms) in Asia, followed by Sri Lanka, mostly with respect to the same three types of platforms noted earlier (goods/products, transport services, tickets and appointments).

In Cambodia, the levels of awareness of various kinds of platforms among Internet users aged 15-65 (Figure 51) had barely translated to use for buying (Figure 52).

Convenience was a key driver of platform use for buying via platforms in India, Sri Lanka and Cambodia, while better pricing was a major driver in Pakistan, and to a lesser extent in India also (Figure 53).

Of the respondents who had used platforms to buy goods or services in the three months leading to the survey, most had made less than five transactions (Figure 54). Heavy use was not widespread.

Among users, many used the platform only to search for goods/ services. Some went onto place an order via the platform, but very few (26% of users in Cambodia, 17% in Sri Lanka, 16% in India and 1% in Pakistan) completed the full transaction (search, order, payment and delivery) through the platform (Figure 55). The reasons for this ranged from lack of a need to do so (they have access to

sufficient offline options), concerns about sharing personal and financial details with third parties, and concerns regarding the quality of the product or whether they would receive the product (Table 11). Not knowing how to use these platforms was also a problem for some, particularly in Cambodia.

Payments were most often done through debit cards (India and Sri Lanka) and cash-on-delivery (all Asian survey countries) among others, with some in India and Pakistan stating they used mobile and/or Internet banking (Figure 56).

Lack of skills was a barrier to greater use of platforms for buying goods and services (Figure 57). In Pakistan, Bangladesh and India, large proportions (87%, 59% and 49% respectively) of those who were aware of these platforms stated that they did not know how to use them. In Nepal and Sri Lanka, the biggest barrier was relevance (45% and 69% respectively).

8 Due to low bases (i.e.: low numbers of people who were aware of platforms as well as low numbers of those that used them), only where the base of users in a particular country is sufficient for meaningful analysis is the data presented.

8

AfterAccess

80

Awareness of platforms for buying/selling (% of Internet users aged 15-65)

Figure 51.

Base Sri L

anka

Indi

a

Paki

stan

Ban

glad

esh

Cam

bodi

a

Nep

al

Internet users 763 919 427 266 804 187

Q Have you heard of opportunities to buy/sell these goods or services through the Internet or apps?

Nepal

7%4%

7% 7%1%

0%

3%

Cambodia

32%

29%

21% 23

%42

%34

%

Bangladesh

24%

19%

23%

10% 13

%16

%

Pakistan

14%

38%

6%4% 3%

India

49%

45% 46

%29

%29

%21

%

Sri Lanka

58%

55%

53%

44%

41%

33%

Goods/products (Amazon, AliExpress, eBay etc.)

Transport/taxi services (Uber, Grab etc.)

Tickets and appointments (movie/railway/doctor appointments etc.)

Accommodation (Airbnb, Booking.com etc.)

Hired help

Microwork/freelance (Upwork, Fiverr etc.)

Asia Report 3.0

81

Use of platforms for buying (% of Internet users aged 15-65 who are aware of platforms)

Figure 52.

NepalCambodiaIndiaSri Lanka

Transport/ taxi services (Uber, Grab etc.)

Tickets and appointments (movie/railway/ doctor appointments etc.)

Hired help

Goods/products (Amazon, AliExpress, eBay) Accommodation (Airbnb, Booking.com etc.)

Base Sri Lanka India Cambodia Nepal

Internet users who are aware of Transport/taxi services (Uber, Grab etc.) 400 396 259 154

Internet users who are aware of Goods/products (Amazon, AliExpress, eBay) 425 430 276 159

Internet users who are aware of Tickets and appointments (movie/railway/doctor appointments etc.) 384 410 185 139

Internet users who are aware of Hired help 307 258 355 141

Internet users who are aware of Accommodation (Airbnb, Booking.com etc.) 321 265 209 88

Q Have you ever bought any of the following goods or services through the Internet or apps?

31%

28%

20%

13%

12%

41%

52%

41%

23% 24

%

9%

13%

10%

9% 9%

11% 13

%

16%

6%

9%

AfterAccess

82

Main reason for buying on platforms (% of Internet users aged 15-65 who had made a purchase/hire via platforms)

Figure 53.

Base Sri L

anka

Indi

a

Paki

stan

Cam

bodi

a

Those who had used platforms for buying 172 239 125 102

Q Why do you choose to use the Internet or apps to search for/buy goods and services?

40%

18%

17%

15%

7%

2%

Sri Lanka

41%

30%

8%11

%11

%

India

11%

54%

10%

11%

2%

Pakistan

4%

17%

39%

8%15

%15

%

Cambodia

None

I can only get these things online where I live (not in shops close by)

Other

Save time since no need to physically go

Quicker service than otherwise

Better prices than otherwise

Convenience of finding what i want in one place

Asia Report 3.0

83

Frequency of buying via platforms during last 3 months(% of Internet users aged 15-65 who had made a purchase/hire via platforms )

Transaction components completed via platforms: buying (% of Internet users aged 15-65 who had made a purchase/hire via platforms)

Figure 54.

Figure 55.

0

1-5

6-10

>10

Search only

Search and place order only

Search, place order and pay

Search, place order, pay and delivery

Base Sri L

anka

Indi

a

Cam

bodi

a

Those who had used platforms for buying 172 239 102

Q How many times have you bought a good or service using the Internet in the last three (3) months?

Sri Lanka CambodiaIndia

37%

30%

16%

17%

68%

29%

2%

Note: The question was administrated differently in Pakistan and is not comparable with the other three countries.

Base Sri L

anka

Indi

a

Paki

stan

Cam

bodi

aThose who had used platforms for buying 172 239 125 102

Q In your most frequent online purchases or hires do you use the Internet to:

Sri Lanka PakistanIndia Cambodia

38%

25%

13%

23%

39%

33%

12%

16%

81%

12%

5%

1%

37%

34%

18%

11%

59%

31%

7%

3%

AfterAccess

84

Sri L

anka

Indi

a

Paki

stan

I don’t need to (e.g.: I can buy all necessary goods/services from physical stores) 33% 31% 25%

I don’t know how to 22% 22% 64%

I am not certain that I will receive the goods/services 21% 20% 2%

I cannot be certain of the quality of the product 20% 23% 3%

Delivery charges are too high 7% 18% 6%

I am not certain that my payment will reach the seller 6% 10% 3%

There is no option to place order or do payment online 6% 13% 1%

It takes too much time 5% 15% 1%

I’m not comfortable sharing personal details online with third parties 4% 24% 4%

I am not comfortable using sellers/service providers that I don’t know 3% 5% 2%

I’m not comfortable sharing financial details online with third parties 2% 11% 0%

Online prices of goods/services are too high 1% 11% 0%

I’ve had a negative experience in the past 0% 7% 0%

I’ve heard of people having negative experiences with these platforms 0% 4% 2%

Q In your most frequent online purchases or hires, what are the reasons you don’t place an order or make a payment through the Internet or mobile apps?

Reasons why platform users stop at search when buying(% of platform users aged 15-65 who did not place orders or pay via platforms)

Table 11.

Base Sri L

anka

Indi

a

Paki

stan

Platform users who did not place orders or pay via platforms 101 175 110

Asia Report 3.0

85

Base Sri L

anka

Indi

a

Paki

stan

Cam

bodi

a

Platform users who purchased through platforms 172 239 125 102

Pakistan

19%

8%10

%33

%

1%

12%

1%

6%6%

6%

22%

27%

5%6%

8%18

%

3%

10%

2%1%

India Cambodia

45%

42%

1%1%

1%2%

8%

31%

30%

13%

11%

8%

3%2%

1%1%

1%

Sri Lanka

COD (cash on delivery)

Debit card

Other

Credit card

Transfer via ATM/Bank

Mobile-banking/Internet-banking

Payment in kind or via exchange of other goods/services

Mobile money transfer/balance transfer

Post office

Online payment (e.g.: Paypal, Paytm, Virtual account)

Western Union

Usual payment method (% of Internet users aged 15-65 who made a purchase/hire via platforms)

Figure 56.

Q What methods of payment do you usually use for your purchases?

AfterAccess

86

59%

26%

3%3%

2%5%

2%

Bangladesh

10%

69%

4%4%

2%11%

Nepal

29%

23%

15%

6%

1%1%

26%

Cambodia

22%

45%

6%8%

5%

2%

13%

Sri Lanka

49%

28%

7%5%

2%2%9%

India

87%

9%

1%2%

Pakistan

Main reason for not buying on platforms (% of Internet users aged 15-65 who are aware of but don’t use platforms for buying)

Figure 57.

BaseSri Lanka India Pakistan Bangladesh Cambodia Nepal

Respondents who are aware of but don't use platforms 172 586 302 227 421 158

Q What is the primary reason you don’t buy goods/services through the Internet or mobile apps?

I don’t know how to

I don’t need to (e.g.: I can buy all necessary goods/ services from physical stores)

I am not certain that I will receive the goods/services

I’m not comfortable sharing personal details online with third parties

Delivery charges are too high

Other I cannot be certain of the quality of the product

Asia Report 3.0

87

There was low use of platforms to sell products or services across all Asian countries surveyed (Figure 58).

The bases of respondents in all the countries, except for India, become too small to analyze meaningfully beyond this point, therefore only the Indian case of use is further considered.

Better job rates/prices and getting access to a larger number of customers were the key drivers for use (Figure 59). Only 27% of sellers in India had sold more than five products or services over the last three months (Figure 60). Only 12% of those who used platforms to sell goods or services completed the full transaction (search, order, payment and delivery) within the platform – many only used the platform to search for and accept jobs (Figure 61). The reasons for this ranged from lack of a need to do so (they have access to sufficient offline options) followed by not knowing how to accept orders and receive payments online (Table 12). Concerns about sharing personal and financial details with third parties and low online

prices of goods or services were also problems for some.

Most of those who used platforms for selling goods or services claimed that what they earn online was not essential but nice to have. However, 24% claimed that the earnings are essential to them (Figure 62).

Across the Asian countries surveyed, the key reason for not using platforms to sell services (among those who were aware of the possibility) was lack of need (Figure 63). The second most common reason was lack of knowledge on how to, especially in Cambodia.

9 Due to low bases (i.e.: low numbers of people who are aware of platforms as well as low numbers of those that use them), only where the base of users in a particular country is sufficient for meaningful analysis is the data presented

PlatformsUsing platforms for selling goods and services9

AfterAccess

88

Use of platforms for selling (% of Internet users aged 15-65 who are aware of platforms)

Figure 58.

Goods/ products (Amazon, AliExpress, eBay)

Tickets and appointments (movie/railway/ doctor appointments etc.)

Accommodation (Airbnb, Booking.com etc.)

Transport/ taxi services (Uber, Grab etc.) Hired help

Base: Internet users who are aware of platforms for: Sri Lanka India Cambodia Nepal

Goods/products (Amazon, AliExpress, eBay) 425 430 276 159

Transport/taxi services (Uber, Grab etc.) 400 396 259 154

Tickets and appointments (movie/railway/doctor appointments etc.) 384 410 185 139

Accommodation (Airbnb, Booking.com etc.) 321 265 209 88

Hired help 307 258 355 141

Q Have you ever sold any of the following goods or services through the Internet or apps?

Sri Lanka

4% 4%

2% 2%

1%

India

12%

14%

9% 9%

12%

Cambodia

1%

2%

0%

1%

3%

Nepal4%

1%

4%

3% 3%

Asia Report 3.0

89

Main reason for selling on platforms (% of Internet users aged 15-65 who sold goods or services via platforms)

Frequency of selling via platforms during last 3 months (% of of Internet users aged 15-65 who sold goods or services via platforms)

Figure 59.

Figure 60.

It offers better job rates/prices than otherwise

It provides access to a larger number of customers than otherwise

I need to be able to control my own schedule due to child care, school, or other obligations

There are not many other job opportunities in my area

To gain work experience for future job opportunities

It helps to fill gaps or fluctuations in other sources of income

Other

For fun, or to do something with my spare time

None

0

1-5

6-10

>10

Base Indi

a

Platform users who sell through platforms 106

Base Indi

a

Platform users who sell through platforms 106

Q What is your main reason for using these websites/apps to sell goods and/or services?

India

India

Q How many times have you sold a good or service (by taking on jobs) using the Internet in the last three (3) months?

29%

39%

34%

25%

19%

15%

12%

12%

8%

4%2%

0%

2%

AfterAccess

90

It is an important component of my budget, but not essential

Base Indi

a

Platform users who sell through platforms 106

Q In the most frequent way you have used to earn money by selling things or taking on jobs through the Internet, do you usually

Q Which of the following statements best describes the income you earn from using these services?

Transaction components completed via platforms: selling (% of Internet users aged 15-65 that sold goods or services via platforms)

Figure 61.

38%

18%

12%

32%

India

Base Indi

a

Platform users who sell through platforms 106

Importance of earnings from platforms (% of Internet users aged 15-65 who sell goods or services via platforms

Figure 62.

It's nice to have but I could live comfortably without it

It is essential for meeting my basic needs

24%

24%

53%

India

Search for jobs/orders only

Search for and accept jobs/orders only

Search for, accept and receive payment for jobs/orders

Search for, accept, receive payment and deliver jobs/orders

Asia Report 3.0

91

Reason

Ind

ia

I don’t need to 50%

I don’t know how to 42%

I’m not comfortable sharing financial details online with third parties 18%

Online prices of goods/services are too low 15%

I am not comfortable using buyers/service providers that I don’t know 14%

There is no option to accept order or receive payment online 13%

I’m not comfortable sharing personal details online with third parties 11%

I am not certain that I will receive payment from the buyer 11%

I’ve had a negative experience in the past 11%

It takes too much time 9%

I’ve heard of people having negative experiences with these platforms 4%

Service provider commission is too high 2%

Other 0%

Q In your most frequent sales, what are the reasons you usually don’t accept the order/payment through the Internet or mobile apps?

Reasons why platform users stop at search when selling (% of platform users aged 15-65 who did not accept jobs or payment via platforms)

Table 12.

Base Indi

a

Platform users who did not accept jobs or payment via platforms 101

AfterAccess

92

I don’t need to

I don’t know how to

I am not certain that I will receive the payment

Charges from the website/app are too high

It takes too much time

Other

84%

8%

3%1%

5%

Sri Lanka

53%

23%

6%5%

4%

10%

India

42%

40%

2%1%

3%

12%

Cambodia

76%

4%1%

2%

17%

Nepal

Reason for not selling on platforms (% of Internet users aged 15-65 who are aware of but don’t use platforms for selling)

Figure 63.

BaseSri Lanka India Cambodia Nepal

Platform users who don't sell through platforms 405 356 506 157

Q What is the primary reason you don’t sell goods or services through the Internet or mobile apps?

Asia Report 3.0

93

Cybersecurity

AfterAccess

94

Cybersecurity awareness Internet users in Sri Lanka and Nepal were asked if they had faced cybersecurity issues related to their devices or their online accounts. Incidence was very low (Figure 64), only 5% and 2% of Internet or social media users aged 15-65 in Sri Lanka and Nepal respectively answered “yes”.

10 Questions related to cybersecurity were included in an updated version of the AfterAccess questionnaire which was used only in the Asian countries where fieldwork took place in the latter part of the survey period (i.e.: Sri Lanka and Nepal).

10

Base Sri L

anka

Nep

al

Internet or social media users 761 713

Q To your knowledge have any of your devices or accounts ever been taken over by someone else, either through the Internet or in person?

Sri Lanka

Nepal

5% 2%

Cyber security (% of Internet or social media users aged 15-65)

Figure 64.

Asia Report 3.0

95

Photo Credits

Adli WahidMuhammad Raufan

Stephen J MasonAnnie Spratt

Shilajit D.C.Malik Earnest

afteraccess.net | @afteraccess | lirneasia.net/afteraccess | @lirneasia