A life satisfaction measure based on judgment theory

H. L. MEADOW, J. T. MENTZER, D. R. RAHTZ AND M. J. SIRGY*

A L I F E S A T I S F A C T I O N M E A S U R E B A S E D O N

J U D G M E N T T H E O R Y

ABSTRACT. A life satisfaction measure is developed in this paper drawn from a judgment-type theory. This measure is based on the theoretical notion of life satisfaction being a function of a comparison between perceived life accomplishments and a set of evoked standards. These standards are classified as to their derivative sources (e.g,, the life accomplishments of relatives, friends, associates, past experience, self-concept of strengths and weaknesses, and average person in a similar position) and different forms (e.g., standards based on ideal, expected, deserved, minimum tolerable, and predicted outcomes). Two studies were conducted on the elderly to ascertain the reliability and validity of the life satisfaction measure. The measure was found to be internally consistent and possessing construct validity correlations through high and positive correlations with the Delighted-Terrible Life Satisfaction Scale; through significant and positive correlations with cognitive age, income, employment, education, marital status, social contact, activity, religiosity, morale, television viewership, and self- rated health; and through nonsignificant correlations with chronological age, gender, and parenthood.

INTRODUCTION

Over the past few decades, a number of researchers f rom various

disciplines have directed considerable attention to the construct labeled

life satisfaction. Interest in life satisfaction has led to numerous attempts

to measure this complex psychological state f rom different perspectives

such as quality of life investigations, examinations of subjective well-

being, research on happiness and morale, and attempts to assess

successful aging. Table I provides a set of examaples of the variety of

measures that have been developed to tap this phenomenon . For a

comprehens ive review of life satisfaction measurement issues, the

reader should see Chamber la in (1988) and Diener (1984), and particu-

larly the works of George and Bearon (1980), Conte and Salamon

(1982), Larson (1978), and McNeil et al. (1986) for comprehens ive

discussions of the use of life satisfaction measures designed specifically for the elderly.

Chamber la in (1988) has suggested that life evaluation assessment

measures can be categorized into four groupings based on major

distinctions between well-being dimensions. The distinction groupings

Social Indicators Research 26:23--59, 1992. �9 1992 KluwerAcademic Publishers. Printed in the Netherlands.

24 H. MEADOW E T A L .

TABLE I Examples of popular life satisfaction measures

Measure Sources

Affect Balance Scale Affect Intensity Measure The Affectometer Delighted-Terrible Scale General Well-Being Schedule Happiness measure Index of General Affect Life Satisfaction Index ILSI] Life Satisfaction Index-A ILSI-A] Memorial University of Newfoundland Scale of

Happiness [MUNSH] Mood Survey

Philadelphia Geriatric Center Morale Scale IPGCMSl

Satisfaction with Life Scale Self-Anchoring Ladder Self-Description Inventory Well-Being Subscale of the Differential

Personality Questionnaire

Bradburn (1969) Larsen (1983) Kammann and Flett (1983) Andrews and Withey (1976) Dupey (1978) Fordyce (1977) Campbell et al. (1976) Neugarten et al. (1961) Wood etal. (1969)

Kozma and Stones (1980) Underwood and Froming

(1980) Lawton (1975); Morris and

Sherwood (1985) Diener etal. (1983) Cantril (1985) Fordyce (1978)

Tellegen (1979).

suggested are: (1) affective (emotional) or cognitive (rational); (2) positive (+) or negative (-); (3) frequency [numbers of times occurring[ or intensity [level of strength associated with each occurrence]; and (4) Inner (self focus) or Outer (outward-directed). These distinctive Chamberlain proposals can be employed to establish a structure for concepts related to subjective well-being.

The theoretical underpinnings for these life satisfaction measures come from a variety of conceptual frameworks, such as the telic theories (e.g., Wilson, 1960), pleasure and pain theories (e.g., Houston, 1981), activity theories (e.g., Csikszentmihalyi and Figurski, 1982), top-down/bottom-up theories (e.g., Kozma and Stones, 1980), associa- tionistic theories (e.g., Schwartz and Clore, 1983), and judgment theories (e.g., Carp and Carp, 1982). As can be seen, there is a virtual plethora of measurement approaches and accompanying theoretical perspectives.

In this paper, the focus will be on judgment theory based conceptual- izations and measurement of the life satisfaction construct. Attention is

L I F E S A T I S F A C T I O N M E A S U R E 25

focused on examining a variety of life satisfaction measures based on judgment theories, assessing measurement adequacy of these judgment based life satisfaction measures, and introducing a new life satisfaction measure based on a particular judgment theory referred to as "congruity theory."

J U D G M E N T T H E O R I E S IN L I F E S A T I S F A C T I O N R E S E A R C H

The central postulate in all judgment theories of life satisfaction is that the degree of life satisfaction experienced by a person at any given moment in time is a direct function of a cognitive comparison between some standard and actual conditions. For example, social comparison theory applied to life satisfaction postulates that individuals use others as bases for developing their personal standards. Hence, if individuals judge themselves as better than others, they will be satisfied with their lives; if they judge themselves as worse than others, they will be dissatisfied (Carp and Carp, 1982; Emmons et al., 1983; Michalos, 1980).

Range frequency theory (Parducci, 1968, 1982) posits the standard used in judging one's current life is the person's own past life. If the individual judges that his/her current life exceeds this standard, the person is expected to be satisfied with life. Conversely, if current life falls below this past standard, then the person will be dissatisfied.

Adaptation-level theory (Brickman et al., 1978) offers the premise that life events can be judged as satisfying or dissatisfying as they actually occur. However, the same life events later lose their capability to generate emotional reaction (life satisfaction/dissatisfaction), because people adapt to positive or negative events. In other words, the standard employed in judging the positivity or negativity of life events shifts in the direction of the perceived event. This movement essentially reduces the discrepancy between the perceived event and the judgment standard employed (Brikman and Campbell, 1971; Cameron, 1974; Feinman, 1978; Wortman and Silver, 1982).

Aspiration-level theory maintains that the degree of life satisfaction/ dissatisfaction experienced is a function of the ratio of fulfilled desires to total desires. Hence, people having unrealistic aspirations usually end up feeling very dissatisfied with their lives because these unrealistic

26 H. M E A D O W E T A L .

aspirations inflate the "total desires" variable, thereby producing a lower ratio of fulfilled desires (Easterlin, 1974; Emmons et al., 1983; Gerrard et al., 1982; Gibbs, 1973; Kammann, 1982; McGill, 1967; Wilson, 1960).

There is some evidence resulting from quality-of-life research dem- onstrating that discrepancies between perceptions of life accomplishments (or outcomes) and life expectancies (or life standards) have consistent and predictable zero-order relationships with life satisfaction (see Michalos, 1985, 1986a, 1986b, for extensive reviews). Michalos (1986b) found that life satisfaction is related significantly to the discrepancy between current experiences and personal standards of comparison in over 80% of the tests provided by the existing research. Such results support the notion of developing a comprehensive and strong measure of life satisfaction based on judgment theory.

Chamberlain (1988) contends this concept of multiple discrepancies "may eventually be shown to be broad general aspects underlying the construct (life satisfaction), but more research needs to be done to establish this" (p. 591). Chamberlain (1988) suggests that these ideas (discrepancies) should provoke further research exploration in order to eliminate the speculative nature of the concept and empirically establish its efficacy. Employing judgment theory derived from the multiple discrepancies concept and empirically tested may provide the vehicle for pulling together the "diverse, fragmentary and suggestive" evidence for such a categorical distinction of well-being measures.

L I F E S A T I S F A C T I O N M E A S U R E M E N T B A S E D ON

J U D G M E N T T H E O R I E S

Most of the judgment studies of life satisfaction have employed "raw" satisfaction measures; that is, self-report type measures of happiness and well-being. Judgment studies which have employed such measures of general affect treat them as dependent measures. The measures examine the relationship between a comparison state (e.g., discrepancy between some standard and actual state) and life satisfaction. That is, these efforts treat the various comparison states as independent variables. None of the judgment studies appears to have employed a measure of life satisfaction based on the generated affect from the


various comparison states. Only the Neugarten et al. (1961) Life Satisfaction Index (LSI) incorporates the concept of a comparison state into the measure, itself; however, even it fails to go "far enough," since it treats the comparison in a unidimensional fashion.

In essence, judgment theories maintain that life satisfaction is directly influenced by a variety of standards. Hence, if a life satisfaction measure is developed which taps the degree of satisfaction experienced through judgments of actual states compared to a set o f specific standards, such a measure may prove to be a fruitful development. The Neugarten et al. (1961) LSI measures the degree of congruence between desired and achieved goals. This measure, of course, is a step in the right direction; however, it lacks in theoretical richness, since the LSI employs a single type of standard/desired goal. Judgment theories make a case for the inclusion of a set of standards because life satisfaction can be affected by a variety of comparison processes involving more than one st.andard.

If a life satisfaction measure is to be developed based on the argument of multiple standards, then what is required is a specific judgment-type theory which establishes the set of standards employed and how the standards are actually aplied in judging personal life circumstances. Such a theory may be found in the works of Sirgy (1980; 1983; 1984a; 1984b; 1985; 1986; 1987a; 1987b) and Sirgy and Tyagi (1986). This program of research is referred to as "congruity theory" and is articulated mainly in consumer psychology -- more specifically, in modeling and explaining consumer satisfaction/dissatisfaction.

A L I F E S A T I S F A C T I O N M E A S U R E B A S E D ON

A J U D G M E N T - T Y P E T H E O R Y

Sirgy's congruity theory is based on the notion that an evaluation process determines satisfaction/dissatisfaction. Evaluation refers to judging the positivity or negativity (emotional enhancement/debasement potential) of an object, person, or event. The extent to which an object, person, or event is positively/negatively evaluated is a function of the congruity between the valence properties of the perceived object/ person/event and the referent beliefs against which the perceived object/person/event is judged. The comparison between perceived


objects/persons/events and referent beliefs produces four possible conditions. These are:

(1) positive congruity (i.e., an object/person/event perceived as not discrepant from positive expectations),

(2) positive incongruity (i.e., an object/person/event perceived as highly discrepant from negative expectations),

(3) negative conguity (i.e., an object/person/event perceived as not discrepant from negative expectations), and

(4) negative incongruity (i.e., an object/person/event perceived as highly discrepant from positive expectations).

It was argued that the outcome of an evaluation process is mediated by the need for emotional enhancement. Based on this motivational tendency, Sirgy argued that the highest degree of satisfaction will be experienced under a positive incongruity condition, followed by positive congruity, negative congruity, and negative incongruity. For example, an individual judging his/her actual life circumstances is predicted to experience the highest degree of life satisfaction if there is a perception of low expectation life accomplishments and the individual perceives that he/she has exceeded his/her expectation. A moderately high degree of satisfaction would be experienced if his expectation is high and perceives that he/she has met his/her expectation. A moderately low degree of satisfaction is predicted under the condition when the individual's expectation and perception of actual accomplishments are both low. Finally, he/she will experience the lowest degree of satisfaction (or most dissatisfaction) given a comparison between a high expectation and low actuality.

The level of satisfaction experienced as a result of congruity has been argued to be moderated by: (1) the strength of both the perceived outcome and the referent state (expectation); (2) the salience or import- ance of the referent state evoked in evaluation; and (3) the type of the referent state (expectation). The third provides the focus of this paper - - the type of referent state or expectancy evoked in evaluation.

Following congruity theory, the referent state evoked in any evaluation process may take a variety of forms. Examples of expectancy types that are employed in evaluation are:

LIFE SATISFACTION MEASURE 29

�9 Expectations based on what one hopes for ideally �9 Expectations based on what one hopes for realistically �9 Expectations based on what one feels one deserves �9 Expectations based on significant others �9 Expectations based on what one predicts will occur �9 Expectations based on what one can minimally tolerate �9 Expectations based on past experience �9 Expectations based on what one is told to expect �9 Expectations based on what is morally acceptable

The list above is not exhaustive, but serves as an illustrative example of possible referent states. In evaluating one's life accomplishments, one needs to ascertain which referent state(s) or standard(s) is (are) commonly used. Therefore, in developing a life satisfaction measure based on congruity theory, one needs to make a case for selecting the appropriate referent states. Additionally, there is a need to provide empirical evidence that each and every type of evaluation about life accomplishments using a specific referent type also contributes significantly to general or overall life satisfaction. This issue will be covered fully later in the paper.

Assuming this continued individual contribution to overall satisfaction, Sirgy suggested that although the evaluation concept is situation- specific, general satisfaction/dissatisfaction directed toward the object, person, or event over time and across situations can be ascertained by summing (or averaging) evaluation outcomes (satisfaction/dissatisfaction states) from this set of comparison states involving different referent states. In the context of life satisfaction, it can, therefore, be argued that general life satisfaction (dispositional and cross-sectional) can be developed through a summative (or average) index, involving a set of judgments about one's life accomplishments using a set of different standards (referent types). If one accepts this argument then the issue changes from one of single versus multiple standards to one of selection of the appropriate standards.

BASIS OF THE MEASURE

At this point, there is a need to both specify and justify the standards to be included in the life satisfaction measure developed in this paper.


Based on a review of the research literature in judgment theory, congruity theory, and life satisfaction in general, a classification shema of referent states (or standards) were found to be commonly used by people (especially the elderly) in evaluating their life accomplishments.

Referent standards involved in evaluation of life accomplishments may involve a source and a form. The s o u r c e of a standard refers to the principal source of information on which the standard is based. The five s o u r c e s of standards in evaluation of life accomplishments are:

(1) a standard based on the life accomplishments of relatives (e.g., parents, brothers, sisters, aunts, uncles, cousins),

(2) a standard based on the life accomplishments of friends and associates,

(3) a standard based on past experience, or what one was in the past,

(4) a standard based on self-concept or perceived strengths and weaknesses, and

(5) a standard based on the average person in a similar position.

The first two standards are justified based on the life satisfaction research conducted in the context of social comparison theory. This research indicates that people (and especially the elderly) compare themselves with significant others in their evaluation of their lives (e.g., Carp and Carp, 1982; Emmons e t al. , 1983; Michalos, 1980). The third standard, that of past experience, is evidenced in Parducci's (1968, 1982) work in range-frequency theory. The model predicts that the greatest happiness occurs for those who have a negatively skewed distribution of past events. In other words, if the person has experienced what he/she feels are a series of negative life events in his/her past, these events may form a standard with which current life events are compared. The fourth standard refers to one's own self-knowledge of resources, capabilities, and deficiencies. Freedman (1978) argued that in the evaluation of life accomplishments, people have the tendency to use internal standards based on their conception of personal values, needs, abilities and resources (cf. Czaja, 1975). Finally, the fifth standard of the average person, drawn from research on social comparisons in life satisfaction, indicates that people, remote in time and space, can be used as standards of comparison if they are made salient.


Eastedin (1974) argued people will become satisfied with their in- comes, given they find themselves to be making more than the average person in similar positions in society (cf. Gerrard et al., 1982).

The form of a standard in the evaluation of life accomplishments is also argued to involve five distinct categories:

(1) a standard based on ideal outcomes, (2) a standard based on expected ("should be") outcomes, (3) a standard based on deserved outcome, (4) a standard based on minimum tolerable outcome, and (5) a standard based on predicted outcome.

The first two standards are justified in light of the research on aspiration-level in life satisfaction (Carp and Carp, 1982). This research indicates that life satisfaction is a direct function of the discrepancy between the person's actual and aspired life conditions. Aspired life conditions may be quite idealistic or quite realistic. Easterlin (1974) showed significant differences in aspirations for income among people. Also, Gibbs (1973) attributed declining life satisfaction of the more fortunate blacks in the U.S. to their rising aspirations. The third standard underscores people's perception of equity/inequity. If a person perceives having worked very hard in life, this person may feel deserving of "much better" than what was received (Carp and Carp, 1982). The fourth standard refers to the possibility people possess thresholds, and when an experience outcome falls below such thresholds, this results in feelings of intense dissatisfaction (Miller, 1977; Sirgy, 1983). Research in cognitive dissonance supports the fifth standard, underscoring the notion that people attempt to predict states of outcome and, hence, use standards based on these predictions in an attempt to avoid cognitive dissonance. Cognitive dissonance has been argued to cause dissatisfaction (Wicklund and Brehm, 1976; Wyer, 1975).

C O N S T R U C T V A L I D A T I O N

The construct validity of the life satisfaction measure developed in this effort (referred to as the congruity life satisfaction scale or CLS) will be assessed by correlating it with a well-established general life satisfaction


measure. More specifically, the following hypothesis will be tested to ascertain the construct validity of the congruity life satisfaction scale.

Hi: General life satisfaction is an aggregate function of a set of life satisfaction states generated by evaluation of life accomplishments against a set of standards.

The standards are those discussed in the previous section. In testing this hypothesis, the general life satisfaction construct will

be operationally represented by the Delighted-Terrible Life Satisfaction (D-T) measure (Andrews and Withey, 1976). The aggregated life saris- faction states will be represented by the newly developed measure of life satisfaction (the Congruity Life Satisfaction scale or CLS).

Furthermore, additional validation of the proposed measure (CLS) will be ascertained by empirical support for the following relationships, specifically addressing the elderly in this study:

H2: There is no relationship between life satisfaction and chronological age among the elderly (Stock et al., 1983).

Ha: There is a negative relationship between life satisfaction and cognitive age among the elderly, in that elderly persons who feel young are more satisfied with their lives than those who feel old (Bengston et al., 1977; Peters, 1971).

H4: There is a posi t ive relationship between life satisfaction and income among the elderly, in that elderly persons who have higher income are more satisfied with their lives than those who have lower income (Alston et al., 1974; Andrews and Withey, 1976; Bortner and Hultsch, 1970; Clemente and Sauer, 1976; Freudiger, 1980; Kimmel et al., 1978; Larson, 1978; Mancini and Orthner, 1980; Riddick, 1980).

Hs: There is no relationship between life satisfaction and gender among the elderly (Andrews and Withey, 1976; Campbell et al., 1976; Goodstein et aL, 1960; Gurin et al., 1960; Olsen, 1980; Palmore and Kivett, 1977; Saner, 1977; Toseland and Rasch, 1979--1980).


H6: There is a negative relationship between life satisfaction and employment in that employed elderly are more satisfied with their lives than unemployed elderly (Campbell et al., 1976; Cohn, 1979; Near et al., 1978 and 1980; Rice etal., 1979 and 1980; Weaver, 1978).

HT: There is a positive relationship between fife satisfaction and education, in that the more educated elderly are more satisfied with their lives than the less educated elderly (Campbell, 1981; Palmore, 1979; Palmore and Luikart, 1972).

Ha: There is a relationship between life satisfaction and marital status, in that married elderly are more satisfied with their fives than unmarried elderly (Larson, 1978; Andrews and Withey, 1976; Glenn, 1975; Glenn and Weaver, 1979).

H9: There is no relationship between life satisfaction and parenthood (having children) among the elderly (Andrews and Withey, 1976; Glenn and McLanahan, 1981; Glenn and Weaver, 1979).

Hi0: There is a positive relationship between life satisfaction and social contact, in that those elderly who have more social contact with others are more satisfied with their lives than those who have less contact (Okun, Stock, Haring, and Witter, 1984).

Hl1: There is a positive relationship between life satisfaction and activity, in that those elderly who are more actively involved in organized activities are more satisfied with their lives than those who are less actively involved (Beiser, 1974; Markides and Martin, 1979; Palmore, 1979; Palmore and Kivett, 1977; Ray, 1979; Riddick, 1980; Sauer, 1977).

H12: There is a positive relationship between life satisfaction and religiosity, in that those elderly who participate more in religious groups are more satisfied with their lives than those who participate less (Clements and Sauer, 1976; Cutler, 1973; Edwards and Klemmack, 1973; Fendiger, 1970; McClure and Loden, 1982).


HI3: There is a positive relationship between fife satisfaction and morale in that those elderly who have higher morale are more satisfied with their lives than those who have lower morale (Anderson, 1977; Czaja, 1975; Drumgoole, 1981; Ginandes, 1977; Higgins, 1978; Kozma and Stones, 1978; Peterson, 1975; Pomerantz, 1978; Reid and Ziegler, 1980; VanCoevering, 1974; Wilson, 1960).

Hi4: There is a positive relationship between life satisfaction and self-rated health, in that those elderly who rate their health highly are more satisfied with their fives than those who rate their health low (Edwards and Klemmack, 1973; Markides and Martin, 1979; Near et al., 1978; Riddick, 1980; Spreitzer and Snyder, 1974; Toseland and Rasch, 1979--1980; Wessman, 1957; Wilson, 1960; Zeglen, 1977).

H15: There is a negative relationship between life satisfaction and television viewership, in that those elderly who watch more television are more dissatisfied with their lives than those who watch less television (Morgan, 1984; Rahtz, Sirgy, and Meadow, 1988, 1989; Richins, 1988).

METHOD

Pretest

The ten-item scale was pre-tested on fourteen senior citizens drawn from the population through convenience sampling. Reliability analysis on the ten:item scale produced an Alpha of 0.90 (Cronbach, 1951). Respondents were asked to provide oral and/or written feedback on the scale. Based on this feedback, adjustments were made to wording on some of the scale's items. The revised instrument was then qualita- tively rechecked on a small group of elderly (n = 4) and cleared for the main data collection.


Sampling

Two major samples were used in this study (referred to as Study One and Study Two). The sample for Study One was 5 000 drawn through s3;stematic random sampling of a mailing list. The mailing list of approximately 20 000 contained the names of households in the Mid- Atlantic area's population with a member 60 years of age or older. The list itself was obtained from a government sponsored agency, which continually updates its records based upon information supplied by the Social Security Administration indicating that someone in a household has reached 60 years of age. The agency only allowed a one-time mailing, precluding a follow-up mailing to minimize nonresponses, and as a result the total response was 16 percent (n = 791). Thirty-nine of the questionnaires were unusable leaving a 15 percent usable response rate (n = 752). Although the 15 percent response rate is not high, it is considered within the expected levels of single mailing consumer surveys (Kanuk and Berenson, 1975). Methods were also employed to assess the impact of response bias. Feber (1981) suggests comparisons between the true population demographic distribution and the returned sample's demographics (e.g., sex and income). Table II compares the demographics from the sample with the Virginia and United States Census.

The sample for Study One was found to be representative of the Virginia and United States elderly populations in gender. However, Study One was also found to possess more widowed and educated elderly.

The sample for Study Two was obtained from a survey sent to 2 130 residents of Albemarle County, Virginia, who were identified as 55+ years of age by the Virginia Center for Health Statistics. A total of 49 questionnaries were returned because the subject was deceased. Of the remaining 2 081 questionnaires, 560 were returned with usable responses, for a response rate of 26.9 percent. Out of the 560 remaining respondents, 31 respondents were further deleted because they indicated a chronological age below 55 years of age. Working under the restrictions of a one-time mailing precluded a second wave.

Table II describes Study Two demographics and compares them with: Virginia Census data, United States Census data, the sample


TABLE II Demographics of the two studies

Demographic Variable

Study One Study Two (.__N z 752) (._N m 529) X (SD) X (SD) or % or %

Sample N onrespondents Census Data X (SD) Virginia US or% % %

Age: 70.453 72.881 (6.820) (6.825)

Gender: male 40.1 female 59.9

Marital status: married 54.6 single/divorced 9.7 widowed 35.7

Employment Status: employed 14.4 not-employed 85.6

Income: 0--999 0.5 1 000--4 999 8.8 5 000--9 999 13.4 10000--14999 14.2 15000--19999 15.0 20 000--24 999 11.5 25 000--29 999 8.7 30 000--34 999 9.0 35 000--39 999 5.4 40 000--44 999 4.9 45 000 and up 8.6

Education: grammar school 8.2 high school 38.7 two year college 18.1 four year college 19.4 graduate school 15.6

71.0

63.5 22.2 43.1 43.8 36.3 77.8 56.9 56.2

53.5 36.1 64.1 64.9 10.9 22.2 10.3 10.9 24.0 41.7 25.5 24.2

13.8 84.9

28.2 a

1 32.3 u 24.8 ~

9.5 d

13.9 86.1

9.5 34.3 31.1 15.3 22.9 32.0 20.2 28.6 18.0 20.4 5.7 10.1 31.3 8.6 8.7

Sources of Census Data -- Statistical Abstract of the U.S., 1979, 1980, 1984, 1986. z <$20 000,

b = $20--40000, c ~ $40--70000, d ~ $30000+

d r a w n fo r S tudy O n e , a n d n o n - r e s p o n d e n t s f o r S tudy T w o . T h e s a m p l e

fo r S tudy T w o was f o u n d to con ta in : m o r e e lde r l y s l ight ly o l d e r in age

t h a n S tudy O n e ; m o r e ma le s t h a n S tudy O n e , Vi rg in ia , and the U n i t e d

Sta tes ; m o r e s i n g l e / d i v o r c e d t h a n S tudy O n e , Vi rg in ia , and the U n i t e d


States; elderly possessing higher income than Study One; and more

educated elderly than Study One, Virginia, and the Uni ted States.

In an effort to measure potential nonresponse bias, a te lephone survey of selected demographics was conduc ted on a r andom sample of

36 elderly who were in the original sampling frame of Study Two, but

did not respond to the questionnaire. Table II shows that Study Two

respondents were different f rom nonrespondents in gender (more

males), in marital status (more married, less single/divorced, and less

widowed), in employment status (more employed), in income (more

income), and in educat ion (more educated).

Operationalizing the Constructs

The new life satisfaction measure -- the congruity life satisfaction scale: The new life satisfaction measure based on congruity theory involves

ten self-report items in which respondents are asked to indicate

responses using a 6-point scale varying f rom "Very Dissatisfied" - - (1)

to "Very Satisfied - - (6). The specific items were phrased as follows:

1. Compared to your LIFETIME GOALS, IDEALS, and WHAT YOU HAD IDEALLY HOPED TO BECOME, how satisfied are you?

2. Compared to what you feel you DESERVE TO HAVE HAPPENED TO YOU CONSIDERING ALL THAT YOU'VE WORKED FOR, how satisfied are you?

3. Compared to the ACCOMPLISHMENTS OF YOUR RELATIVES (parents, brothers, sisters, etc.), how satisfied are you?

4. Compared to the ACCOMPLISHMENTS OF YOUR FRIENDS AND ASSOCIATES, how satisfied are you?

5. Compared to the ACCOMPLISHMENTS OF MOST PEOPLE IN YOUR POSITION, how satisfied are you?

6. Compared to WHERE YOU'VE BEEN AND HOW FAR YOU HAVE COME ALONG (the progress you have made, the changes you have gone through, or the level of growth you have experienced), how satisfied are you?

7. Compared to WHAT YOU HAVE EXPECTED FROM YOURSELF ALL ALONG CONSIDERING YOUR RESOURCES, STRENGTHS AND WEAKNESSES, how satisfied are you?

8. Compared to WHAT YOU MAY HAVE PREDICTED ABOUT YOUR- SELF BECOMING, how satisfied are you?

9. Compared to WHAT YOU FEEL YOU SHOULD HAVE ACCOM- PLISHED SO FAR, how satisfied are you?

10. Compared to WHAT YOU FEEL IS THE MINIMUM OF WHAT ANYONE IN YOUR POSITION SHOULD HAVE ACCOMPLISHED (AND BE ABLE TO ACCOMPLISH), how satisfied are you?

The following instructions were also included in the measure:

38 H. MEADOW ETAL.

In this section, we would like for you to judge your feelings or level of satisfaction (dissatisfaction) toward YOURSELF AND YOUR LIFE ACCOMPLISHMENTS. Below are a set of questions involving your judgment of yourself and your life accom- plishment using different sets of standards. Please circle the number that shows the level of agreement with the corresponding statement. Number 1 means the strongest level of dissatisfaction and number 6 means the strongest level of satisfaction.

1-Very dissatisfied 3-Somewhat dissatisfied 5-Satisfied 2-Dissatisfied 4-Somewhat Satisfied 6-Very Satisfied

W h a t fo l lowed was a br ie f desc r ip t ion of the i tems des igned to tap a

pe r son ' s eva lua t ion of:

Item 1 - one's life accomplishments involving a standard based on ideal outcomes,

Item 2 -- the standard based on deserved outcomes, Item 3 -- the standard based on relatives, Item 4 -- the standard based on friends/associates, Item 5 -- the standard based on average persons in similar positions, Item 6 -- the standard based on past experience, Item 7 -- the standard based on self-concept of strengths and weaknesses Item 8 -- the standard based on predicted outcomes Item 9 -- the standard based on expected outcomes, and Item 10 -- the standard based on minimum tolerable outcomes.

A n overa l l score of genera l life sat isfact ion was c o m p u t e d for each

r e s p o n d e n t by summing indiv idual sat isfact ion scores ( i tems 1 th rough

10) and d iv id ing by the n u m b e r of i tems (10). Descr ip t ive statist ics of

the measu res a re r e p o r t e d in Tab le III. The mean ref lected a m o d e r -

a tely sat isf ied g roup of r e sponden t s (Study One: X = 4.762, SD -

1.002; S tudy Two: X = 4.854, SD --- 0.850). A l l of the indiv idual

eva lua t ions involving s t anda rds based on relat ives, mos t peop le , and

pas t exper i ence p r o d u c e d the highest sat isfact ion rates in bo th samples .

Eva lua t ions involving dese rved and expec ted s t andards p r o d u c e d the

lowest sat isfact ion scores in bo th samples .

The rel iabi l i ty and val idi ty results of the newly d e v e l o p e d me a su re of

life sa t isfact ion a re r e p o r t e d in the results section.

Measure of general life satisfaction: The mea su re of genera l life satisfac-

t ion se lec ted for test ing the c r i t e r ion val idi ty of the new life sat isfact ion

measu re was the De l igh ted -Ter r ib l e (D-T) Sat isfact ion Scale ( A n d r e w s

and Withey , 1976). T h e D - T Sat isfact ion Scale was r e p o r t e d to have

t e m p o r a l rel iabi l i ty, 0.66 for a 15 minute re l iabi l i ty and 0.40 for a six

m o n t h rel iabi l i ty (Stock, Okun, Har ing , and Wit ter , 1983). A n d r e w s

L I F E S A T I S F A C T I O N M E A S U R E

TABLE Ill Descriptive statistics of socio-psychological measures

39

Construct /terns

Study One Study Two

( N = 752) (_N_=529) X SD X SD

Congruity Life Satisfaction (CLS)

Overall Life Satisfaction (DT)

Cognitive Age

Parenthood

Social Contact

Sat on IDEAL standard Sat. on DESERVED standard Sat. on RELATIVE standard Sat. on FRIENDS standard Sat. on MOST PEOPLE standard Sat. on PAST EXPERIENCE standard Sat. SELF-CONCEPT standard Sat. on PREDICTED standard Sat. on EXPECTED standard Sat. on MINIMUM standard

OVERALL

Alpha =

D-T Standard (time 1) D-T Standard (time 2)

OVERALL

Corr. =

I FEEL as though I am in m y . . . I LOOK as though 1 am in m y . . . I DO most things as though I a m . . . My INTERESTS are those o f . . .

OVERALL

Alpha =

Combined scores 3 item dichotomy measuring children, grandchildren and pets (1 = yes, 2 = no)

(+) Children and grandchildren are important in life

(+) Our family is a close knit group (+) My friends are a big part of my life (+) Friends often come to me for advice (+) I have an active social life

OVERALL

Alpha

4.701 1.079 4.780 1.095 4.556 1.121 4.781 1.096 4.907 1.000 5.000 1.050 4.827 0.969 4.837 1.002 4.953 0.929 4.931 0.954 4.989 0.896 4.971 0.982 4.805 0.994 4.778 1.022 4.656 1.070 4.741 1.107 4.516 1.132 4.640 1.086 4.705 1.109 4.918 1.003

4.762 1.002 4".854 0.850

0.950 =0.952

4.586 1.047 4.530 1.023

4.508 0.999 5.742 0.873

0.87

3.477 1.004 3.825 1.022 3.584 0.772 3.082 0.804 3.322 1.086 3.677 1.038 3.175 0.883 3.431 1.049

3.424 0.920 3.727 0.831

0.87 =0.891

4.02 0.525

5.570 0.816

4.899 1.126 4.919 1.030 4.026 1.168 3.973 1.284

4.709 0.819

0.73


Table III (continued)

Study One Study Two

(N = 752) (N = 529) Construct Items X SD X SD

Activity ( - ) I would rather spend a quiet evening 3.498 1.276 at home

( - ) Television is my primary form of 3.839 1.585 entertainment

(--) I stay home most evenings 4.597 1.235

OVERALL 3.505 0.941 Alpha ~ 0.73

Religiosity (+) I am an active member in my church 4.231 1.726 or temple

Self-Rated How would you rate your health at the 1.821 0.734 2.969 0.706 Health 1 present time

"4" excellent, "3" good, "2" fair, "1" poor

Self-Rated (+) I am in good physical condition 4.423 1.290 Health 2

(+) I work hard at staying in good 4.215 1.106 physical shape

(+) Being active is a good way to to stay 5.496 0.611 healthy

(4-) I pay close attention to my health 4.745 0.953 (+) I exercise regularly 4.013 1.470 (4-) I pay close attention to my diet 4.184 1.274

OVERALL 4.500 0.847 Alpha ~ 0.77

Morale ( - ) Everything is changing too fast 3.040 1.365 ( - ) My opinions on things don't count 3.873 2.086

very much ( - ) Not many people respect senior 3.678 1.362

citizens ( - ) Ifeelold 4.590 1.461 ( - ) I often find myself feeling lonely 4.302 1.567 ( - ) Other people I talk to seemto know 4.356 1.284

more about things than I do (--) I often think about my personal safety 2.423 1.215

OVERALL 3.298 0.826 Alpha ~ 0.63

TVViewership (+) Television is my primary form of 3.327 1.712 entertainment

Frequency ( # of hours/wk) 18.57 15.65

All items are scaled as 6-point forced choice Likert-Type Scales, except for "Self-Rated Health 1" which is a 4-point scale of (1) excellent, (2) good, (3) fair, (4) poor.


and Withey (1976) reported high convergent validity with other self- report measures of life satisfaction and nomological validity by providing empirical support for relationships between life satisfacation and external variables such as self-efficacy, marriage, and standard of living. Other positive and strong evidence of the reliability and validity of the D-T measure were reported by Larsen, et al. (1983).

The measure is a one question self-report measure, asked twice within the instrument, with the responses coded using a 9-point scale varying from "Delighted" (1) to "Terrible" (7), with two additional categories for "Neutral" (neither satisfied nor dissatisfied) and "I Never Thought About It" which were treated as missing values in the statistical analyses. The categories of the scale varying from "Terrible" to "Delighted" are: "Terrible," "Pleased," "Mostly Satisfied," "Mixed (About Equally Satisfied and Dissatisfied)," "Mostly Dissatisfied," "Un- happy," and "Terrible." The precise item wording was phrased as follows: "How do you feel about your life as a whole?"

In Study One, the D-T scale was inserted once at the beginning of the questionnaire and another time close to the questionnaire's end. Correlating these two identical items is viewed as providing an indica- tion of a 20-minute temporal reliability. The correlation between these two items in Study One was 0.837, indicating high temporal reliability of the D-T scale. Further statistical analysis involving the D-T scale was conducted by summing the two items producing one D-T score for each respondent. Study Two included the D-T measure in the questionnaire only once (placed around the middle of the instrument, not adjacent to the CLS items). Descriptive statistics of the D-T scale are reported in Table III.

Operationalization of the other comparative measures: Eight other socio-psychological constructs, seven demographic constructs, and one behavioral construct were included in the questionnaire. The eight additional socio-psychological constructs were: cognitive age, parenthood, social contact, activity, religiosity, two measures of self-reported health, and morale. Seven demographic eonstructs were employed: chronological age, income education, gender, religious affiliation, and marital status. The behavioral construct was television viewership. Tables II and III provide summaries of the descriptive statistics of the

42 H. MEADOW ETAL.

socio-psychological, demographic, and behavioral constructs. The socio-pyschological items were either borrowed directly from the literature (see the hypotheses section) or developed for this study based on the available literature. For example, the cognitive age scale (Barak and Schiffman, 1981) and the first self-rated health scale (Maddox and Douglas, 1973) were taken directly from the literature. The morale scale, the second self-rated health scale, and the other measures were developed from relevant issues provided by a review of related literature.

RESULTS

In this section, the results of the statistical analysis are reported. First, the results pertaining to the internal consistency structure of the life satisfaction measure are provided. Second, the results pertaining to the construct validity of the measure (Hypothesis 1 through Hypothesis 14) are presented.

Internal Consistency Structure

Although each item in the Congruity Life Satisfaction (CLS) measure represents a different theoretical construct (evaluation of life accomplishments against a specific type standard), it was argued that the aggregation of these evaluations make up the general life satisfaction construct. Hence, since the focus is on the aggregate measure (i.e., the sum of the individual evaluations of life accomplishments based on standard 1 through 10, as reflected in items 1 through 10), it is expected that most of the individual evaluations will be consistent with one another. As a result, an internal consistency test should be the appropriate procedure employed. A Cronbach Alpha analysis was conducted on the CLS scale, with the results shown in Table IV for both studies. As can be noted by examining Table IV, the CLS scale shows high internal consistency (for Study One 0.946 and for Study Two 0.951 Alpha coefficients) attesting to the reliability of the CLS measure.

O

+3

r~

r~

C

C


L ~ . ~

e,

N

V ~V ~ ~= =m~=

o~-- o~

" ~ ~ S " . . . . . . . . . .

+ o , . . . . . . o m m m m + + + m m ~ . u m= " u m .~ I II II II II n II II II II

44 H. MEADOW ETAL.

Analysis Pertaining to Construct Validity

Table V shows a list of correlations between the Delighted-Terrible

(D-T) measure and the CLS measure (as well as its individual components). The correlation pattern is generally high, which provides support for the construct validity of the CLS measure (r = 0. 470 for Study One and 0.394 for Study Two).

TABLE V Correlations between the congruity life satisfaction scale and the delighted-terrible scale

Study One Study Two (N = 752) (N = 529)

OVERALL

CLS (1) 0.375 0.372 CLS (2) 0.431 0.361 CLS (3) 0.341 0.310 CLS (4) 0.359 0.326 CLS (5) 0.378 0.389 CLS (6) 0.369 0.366 CLS (7) 0.368 0.346 CLS (8) 0.376 0.333 CLS (9) 0.357 0.333 CLS (10) 0.338 0.316

CLS 0.470 0.394

Notes: All correlations are significant with p < 0.0001. D-T scores were computed by combining the two D-T items in STUDY ONE only. Legend: See Table IV.

With respect to the correlations between each CLS item and D-T, it

was found that all of the items seem to be equally correlated with D-T. The pattern of correlations suggests that global life satisfaction may be more affected by evaluations involving all of the included standards.

Relationship between life satisfaction and chronological age (H2): Ac- cording to the research literature, no relationship exists between life satisfaction and chronological age. The correlation between CLS and chronological age was - 0 . 0 3 in Study One (p > 0.10) indicating no relationship; however, a significant positive relationship was found in Study Two (r ---- 0.151, p < 0.01). This finding provides only modest


support for the validity of the CLS scale. The correlation between D-T and chronological age produced a nonsignificant correlation (r ~- 0.01, p > 0.10 for Study One and r = 0.065, p > 0.10 for Study Two) indicating no relationship, contributing to the nomological validity of the CLS scale (see Table VI). It can be argued that the unexpected moderately low correlation between chronological age and CLS (shown in Study Two) may be attributed to the "noise" in the data, since the D-T measure failed to produce a similar correlation.

Relationship between life satisfaction and cognitive age (//3): The correlation between CLS and cognitive age was -0.171 (p < 0.01) and -0 .029 (p > 0.10) for Study One and Study Two, respectively. Only Study One supported the expected negative relationship between life satisfaction and cognitive age. More specifically, the correlation for Study One indicates that those elderly who feel younger than their chronological age are more satisfied with their lives than those who feel their age (or older). These findings all together provide only moderate support for the validity of the CLS scale (see Table VI).

The correlation between D-T and cognitive age was -0 .330 (p < 0.001) for Study One and -0 .227 (p < 0.01) for Study Two. These are stronger correlations than those found between CLS and cognitive age. This finding reinforces the existence of the relationship beteen life satisfaction and cognitive age as hypothesized; a finding which, in turn, provides only little to moderate support for the validity of the CLS scale (see Table VI).

Relationship between life satisfaction and income (H4): According to the life satisfaction literature, a relationship between life satisfaction and income should be expected. The correlation between CLS and income was 0.257 (p < 0.001) for Study One and 0.229 (p < 0.01) for Study Two, indicating a relationship. That is, those elderly who have higher income were found to be more satisfied with their lives than those who have lower income. This finding supports the validity of the CLS scale (see Table VI).

The correlation between D-T and income was found to be 0.210 (p < 0.001) for Study One and 0.225 (p < 0.001) for Study Two, which reinforces the existence of a relationship between life satisfaction

4 6 H. M E A D O W E T A L .

"V-.

"T,

0 "8

e- e'.,

f.I.1

m.~,. a

E , -

",~_,, t~

0

~. ~ .

I I I I I I I I I I

oeao r e,l

T I I I

- - - - ~ e - !

_ - ~ -. ~, # ~ -~'~

A - ~ ~ o~' ~-- ~- ~ ~ ~ ~ ~ tth ~

_ ~ ~.~ . ~ u~, ~ t ~ ~ : ~ , ~ . -~

d

F=

Z 0 >-

P e

_e

{

�9 ~

v _ ~

LIFE S A T I S F A C T I O N M E A S U R E

I I I I I I I I I I I ~ 1 ~ 1 I I I I I I I I I I I

~ o ~ ~ * ~* ~ * * *

47

o

t 'q

, ._t

T

~ E

v V ~

48 H. MEADOW ETAL.

and income, a finding which (as has been stated) is highly prevalent in life satisfaction studies. This finding provides additional support for the validity of the CLS scale (see Table VI).

Relationship between life satisfaction and gender (/45): According to the life satisfaction literature, no relationship exists between life satisfaction and gender among the elderly. Table VI shows nonsignificant correlations between CLS and gender in both studies (Study One, r = -0.033, p > 0.05; Study Two, r ---- 0.028, p > 0.05), reinforcing the validity of the CLS measure. However, it should be noted that a significant correlation was detected between D-T and gender in Study One (r -- -0.173, p < 0.05), but not in Study Two (r -- -0.078, p > 0.10). More specifically, male elderly respondents were found to be more satisfied with their lives than their female counterparts. Nevertheless, this difference between male and female elderly seems to be highly negligible, since it was not replicated by Study Two.

Relationship between life satisfaction and employment (H~): Life satisfaction literature suggests a negative relationship between life satisfaction and employment exists. That is, employed elderly report more satisfaction than unemployed elderly. The correlations shown in Table VI do not support H 6 (Study One, r = -0.022, p > 0.05; Study Two, r = 0.033, p > 0.05). Similar results were obtained using the D-T measure.

Relationship between life satisfaction and education (//7): Based on previous life satisfaction research, it was hypothesized that there would be a positive relationship between life satisfaction and education. That is, those elderly found to have higher educational levels report greate~ satisfaction with their life accomplishments than those who have lower educational levels. The correlation between CLS and education was 0.104 (p < 0.01) for Study one and 0.124 (p < 0.01) for Study Two. Although these correlations are rather small, they nevertheless provide support for the hypothesis and provide additional support for the validity of the CLS scale (see Table VI).

The correlation between D-T and education was found to be 0.143, (p < 0.01) for Study One and 0.153 (p < 0.01) for Study Two,


findings which are consistent with the correlation between CLS and education in previous studies. The correlation reinforces the existence of the relationship between life satisfaction and education, and therefore, reinforces the validity of the CLS measure (see Table VI).

Relationship between life satisfaction and marital status (H8): According to the research literature a relationship between life satisfaction and marital status should exist. That is, married elderly report higher satisfaction than unmarried ones. Table VI shows significant correlations for Study One (r ~ -0.131, p < 0.01), but not for Study Two (r = 0.008, p > 0.05). However, the results employing the D-T measure produced the expected finding in both studies (Study One, r = - 0 . 1 7 3 , p < 0.01; Study Two, r = - 0 . 1 0 9 , p < 0.05).

Relationship between life satisfaction and parenthood (/-/9): Life satisfaction literature suggests there is no relationship between life satisfaction and parenthood among the elderly. Table VI shows nonsignificant correlations between life satisfaction (CLS and D-T measures) and parenthood (Study One, r - 0.052, p > 0.05 with respect to the CLS measure and r -- 0.018, p > 0.05 with respect to the D-T measure). Study Two did not have a measure of parenthood.

Relationship between life satisfaction and social contact (/-/10): Previous life satisfaction research has demonstrated a positive relationship between life satisfaction and social contact. That is, those elderly who have more social contact with others are expected to report higher life satisfaction than those who have less social contact. Table VI shows significant correlations between life satisfaction and social contact (Study One, r - 0.309, p < 0.01 with respect to the CLS measure and r ~ 0.365, p < 0.01 with respect to the D-T measure). Study Two did not have a measure of social contact.

Relationship between life satisfaction and activity (H1/): According to previous life satisfaction research, there is a positive relationship between life satisfaction and activity. That is, those elderly who are more actively involved in organized activities are more likely to be satisfied than those who are less actively involved. Table VI reports

50 H. MEADOW ETAL.

significant correlations between life satisfaction and activity (Study One, r ~- 0.246, p < 0.01, with respect to the CLS measure and r = 0.294, p < 0.01 with respect to the D-T measure). Study Two did not have a measure of activity.

Relationship between life satisfaction and religiosity (/-/12): Previous life satisfaction research has indicated a positive relationship between life satisfaction and religiosity. More specifically, it is expected that those elderly who report greater participation levels in religious groups, would be more satisfied than those who report less participation. The correlations from Study One in Table VI between life satisfaction and religiosity support this hypothesized relationship (CLS: r ---- 0.187, p < 0.01; D-T: r -- 0.204, p < 0.01). Study Two did not have a measure of religiosity.

Relationship between life satisfaction and morale (Ht3): The correlation between CLS and Morale was -0 .333 (p < 0.001) for Study One indicating that those elderly possessing high morale are more satisfied with their lives than those with low morale. This finding provides support for the validity of the CLS measure. Furthermore, from Study One, the D-T measure was found to significantly correlate with morale in the expected direction (r = -0 .428, p < 0.001). Study Two did not have a measure of morale.

Relationship between life satisfaction and self-rated Health (/-/14): As noted in the method section, two measures of self-rated health, a self- report measure based on a staple-type scale (method 1) and another based on a Likert-type scale (method 2) were employed. With respect to the measure based on the staple-type scale (method 1), the correlation between Self-Rated Health and CLS was 0.251 (p < 0.001) [D-T; r ---- 0.472, p < 0.01] for Study One, and 0.272 (p < 0.001) for Study Two [D-T; r ~ 0.493, p < 0.01]. That is, those elderly who rate themselves as possessing good health were found to be more satisfied with their lives than those who rate themselves as possessing poor health. This finding provides nomological validity to the CLS scale.

With respect to the self-rated health measure based on the Likert- type scale (method 2), again a significant positive relationship between


health and life satisfaction was found in Study One (CLS; r = 0.202, p < 0.001; D-T; r -- 0.237, p < 0.001). Study Two did not have a measure of self-rated health based on method 2.

Relationship between life satisfaction and television viewership (//15): Past research indicates a negative relationship between television viewership and life satisfaction among the elderly. That is, the higher the television viewership the lower the life satisfaction. Both studies produced significant correlations supporting the hypothesized relationship between television viewership and life satisfaction (Study One -- CLS: r = - 0 . 1 5 5 , p < 0.001; D-T: r - - - 0 . 2 2 9 , p < 0 .001 /S tudy Two -- CLS: r = -0 .071, p < 0.10; D-T: r = --0.083, p < 0.05).

DISCUSSION

A study was conducted to develop a life satisfaction measure based on a judgment-type theory. The measure was based on the theoretical notion that life satisfaction is an evaluation involving one's life accomplishments against a set of standards. These standards are thought to involve two types: (1) standards reflecting the form of expectations (ideal, expected, deserved, predicted, and minimum tolerable) and (2) standards reflecting the source of expectations (standards based on one's perception of progressive life accomplishments; based on one's perception of personal strengths and weaknesses -- what one is capable or not capable of having accomplished; and based on one's perception of the average person in a similar position). The measure involved ten self-report items in which subjects were asked to indicate their satisfaction/dissatisfaction responses to items such as, "compared to your lifetime goals, ideals, and what you had ideally hoped to become, how satisfied are you? ... . Compared to what you feel you deserve to have happened to you considering all that you've worked for, how satisfied are you?"

Two studies were conducted using samples of 752 and 529 elderly respondents. Mall surveys were employed. The overall pattern of results provided support for (1) the reliability of both CLS and D-T scales, and (2) the construct validity of both CLS and D-T measures (see Table VII).

52

&

<I , . - i

@

@

@

@

H. M E A D O W E T A L .

,.= , ~ "~ = .,= ..1=

= = a ~ = " = ~ = ' - ~ . - = :;,-. ~ = -

.~ =~ "= S o "5 ~ = " = " = ' = ~ =

. - . - ~ ~ ~..

' - o - ~ . o . - , - , ~ .$ " ~ ~ ~ . 0 . s _=

- - . = ~ ~ . . = < . o ~ _ .= > > > . = .,~

. = ~ ' - ~ - . ' - ~ , ' - ' " o ' - - . ~ ~ , . . . . . . ' - ~ .

�9 '_ .~, .~ .~ -~-, .,~ -~_ -~ -~, " - " - "b. "'~ "~ "~

,.=

c

o.. o #z

~ Z


It should be noted that the study has several limitations. One, the response rate of Study One's mail survey was a low 16 percent. However, given that the method was restricted to the use of a one-time mailing with no follow-up, the response rate may be somewhat acceptable. Also, a comparison between the sample population and the population at large, along two demographic variables (sex and income), indicates the sample to have a similar demographic profile to the elderly population at large. The response rate in Study Two was much higher (26.9 percent) than Study One (16 percent), but comparisons with the nonrespondent sample gave evidence of the presence of response bias given differences found in age, income, employment, marital status, and education. It could be argued that the hypotheses relating life satisfaction to chronological age (H2) cognitive age (H3),

and marital status (Hs), may have been moderately impacted by the presence of such sample bias. As for the hypothesized relationship between employment and life satisfaction (H6), the presence of such bias may have masked the presence of this hypothesized relationship, such that none of the operationalized measures were found to be supportive.

Although it was argued that the reliability of the CLS measures could be tested via an internal consistency-type procedure and the results indicated high internal consistency, future research may focus upon developing multiple indicators of each of the individual evaluation subconstructs of the CLS measure and attempt internal consistency tests of the subconstructs. Furthermore, temporal reliability testing should be attempted, as well.

Future research may further generate testable hypotheses explaining the personality and situational factors affecting the use of the varied standards in life satisfaction decisions. That is, under what circumstances (and for what people) certain standards are more used in global (life) evaluations than others. The results pertaining to the individual evaluations reported in this effort may provide the background for theory formulation.

N O T E

* The authors names appear in alphabetical order. All authors contributed equally to this article.


R E F E R E N C E S

Alston, J. P., Lowe, G. D., and Wrigley, A.: 1974, 'Socio-economic correlates for four dimensions of self-perceived satisfaction', Human Organization 33, pp. 99--102.

Anderson, M. R.: 1977, 'A Study of the Relationship between Life Satisfaction and Self-concept, Locus of Control Satisfaction with Primary Relationships, and Work Satisfaction', Doctoral dissertation, Michigan State University, Dissertation Ab- stracts International 38, 2638 9A (University Microfilms No. 77--25,214).

Andrews, F. M. and Withey, S. B.: 1976, Social Indicators of Well-being: America's Perception of Life Quality (Plenum Press, New York).

Barak, B. and Schiffman, L.: 1981, 'Cognitive age: a nonchronological age variable', in K. B. Monroe (Ed.), Advances in Consumer Research (Association for Consumer Research: Ann Arbor, MI), 8, pp. 602--6.

Beiser, M.: 1974, 'Components and correlates of mental well-being', Journal of Health and Social Behavior 15, pp. 320--7.

Bengston, V. L., Kaschau, P. L., and Ragan, P. K.: 1977, 'The Impact of Social Structure of Aging Individuals', in J. S. Birren and K. W. Schaie (eds.), Handbook of the Psychology of Aging (Van Nostrand Reinhold Company, New York), pp. 327--59.

Bortner, R. W. and Hultsch, D. F.: 1970, 'A multivariate analysis of correlates of life satisfaction in adulthood', Journal of Gerontology 25, pp. 41--7.

Bradburn, N. M.: 1969, The Structure of Psychological Well-being (Aldine, Chicago). Brickman, P. and Campbell, D. T.: 1971, 'Hedonic relativism and planning the good

society', in M. H. Apply (ed.) Adaptation Level Theory: A Symposium (Academic Press, New York), pp. 287--302.

Brickman, P., Coates, D., and Janoff-Bulman, R.: 1978, 'Lottery winners and accident victims: is happiness relative?' Journal of Personality and Social Psychology 36, pp. 917--27.

Cameron, P.: 1974, 'Social stereotypes: three faces of happiness', Psychology Today 8, pp. 63--4.

Campbell, A.: 1981, The sense of well-being in America: recent patterns and trends (McGraw-Hill, New York).

Campbell, A., Converse, P. E. and Rodgers, W. L.: 1976, The Quality of American life (Russel Sage Foundation, New York).

Cantril, H.: 1965, The Pattern of Human Concerns (Rutgers University Press, New Brunswick, N J).

Carp, F. M. and Carp A.:1982, 'Test of a model of domain satisfaction and well being: equity considerations', Research on Aging 4, pp. 503--22.

Chamberlain, K.: 1988, 'On the structure of well being', Social Indicators Research 20, pp. 518--604.

Clemente, F. and Saner, W. J.: 1976, 'Life satisfaction in the United States', Social Forces 54, pp. 621--31.

Cohn, R. M.: 1979, 'Age and the satisfactions from work', Journal of Gerontology 34, pp. 264--72.

Conte, V. C. and Salamon, M. J.: 1982, 'An objective approach to the measurement and use of life satisfaction with older persons', Measurement and Evaluation in Guidance 15, pp. 194--200.

Cronbach, L. J.: 1951, 'Coefficient alpha and the internal structure of tests', Psychome- trika 16, pp. 297--334.

Csikszentmihalyi, J. and Figurski, T. J.: 1982, 'Self-awareness and aversive experience in everyday life', Journal of Personality 50, pp. 15--24.

Cutler, S. J.: 1973, 'Voluntary association participation and life satisfaction: a cautionary research note', Journal of Gerontology 28, pp. 96--100.


Czaja, S. J.: 1975, 'Age differences in life satisfaction as a function of discrepancy between real and ideal self-concepts', Experimental Aging Research 1, pp. 81--9.

Diener, E.: 1984, 'Subjective well-being', Psychological Bulletin 95, pp. 542--75. Diener, E., Emmons, R., Larsen, R., and Griffin, S.: 1983, The Satisfaction With Life

Scale, Working Paper, University of Illinois at Urbana-Champaign. Drumboole, W. P.: 1981, 'Self-concept and life satisfaction as perceived by young,

middle-aged, and senior Adults', Doctoral Dissertation, East Texas State University, Dissertation Abstracts International 41, 2939A. (University Microfilms No. 80-27, 666).

Dupuy, H. J.: 1978, 'The research edition of the general psychological well-being schedule', Unpublished manuscript.

Easterlin, R. A.: 1974, 'Does economic growth improve the human lot? some empirical evidence', in P. A. David and M. W. Reder (Eds.), Nations and Households in Economic Growth (Academic Press, New York), pp. 89--125.

Edwards, N. J. and Klemmack, D. L.: 1973, 'Correlates of life satisfaction: A re- examination', Journal of Gerontology 28, pp. 497--502.

Emmons, R. A., Larsen, R. J. Levine, S., and Diener, E.: 1983, 'Factors predicting satisfaction judgments', Paper presented at the meeting of the Midwestern Psycho- logical Association Convention, May, Chicago.

Feinman, S.: 1978, 'The blind as ordinary people', Journal of Visual Impairment and Blindness 72, pp. 231--38.

Ferber, R.: 1981, 'The role of response rates in evaluating manuscripts for publication', in K. B. Monroe (Ed.), Advances in Consumer Research (Ann Arbor, MI: Associa- tion for Consumer Research) 8, pp. 31--4.

Fordyce, M. W.: 1977, 'The happiness measures: A sixty-second index of emotional well-being and mental health', Unpublished manuscript.

Fordyce, M. W.: 1978, 'Prospectus: the self-description inventory', Unpublished manuscript.

Freedman, J.: 1978, Happy People: What Happiness Is, Who Has It and Why (Harcourt Brace Jovanovich, New York).

Freudiger, P. T.: 1980, 'Life satisfaction among American women', Doctoral dissertation, North Texas State University, Dissertation Abstracts International 40, 6438A. (University Microfilms No., 80--12, 882).

George, L. K. and Bearon, L. B.: 1980, Quality of Life in Older Persons: Meaning and Measurement (Human Sciences Press, New York).

Gerrard, C. K., Reznikoff, M., and Rildan, M.: 1982: 'Level of aspiration, life satisfaction and locus of control in older adults', Experimental Aging Research 8, pp. 119-- 21.

Gibbs, B. A. M.: 1973 'Relative deprivation and self-reported happiness of Blacks: 1946--1966', Doctoral dissertation, The University of Texas, Dissertation Abstracts International 34, 885A. (university Microfilms No. 73--18, 429).

Ginandes, C. S.: 1977, 'Life satisfaction and self-esteem values in men of four different socio-economic groups', Doctoral dissertation, Boston University, Dissertation Abstracts International 38, 1880B. (university Microfilms No. 77--21,590).

Glenn, N. D.: 1975, 'The contribution of marriage to the psychological well-being of males and females', Journal of Marriage and the Family 37, pp. 594--600.

Glenn, N. D. and McLanahan, S.: 1981, 'The effects of offspring on the psychological well-being of older adults', Journal of Marriage and the Family 43, pp. 138--50.

Glenn, N. D. and Weaver, C. N.: 1979, 'A note on family situation and global happiness', Social Forces 57, pp. 960--7.

Goodstein, J., Zautra, A., and Goodhart, D.: 1982, 'A test of the utility of social indicators for behavioral health service planning', Social Indicators Research 10, pp. 273--95.


Gurin, G., Veroff, J., and Feld, S.: 1960, Americans View Their Mental Health (Basic Books, New York).

Higgins, D. H.: 1978, 'Self-concept and its relation to everyday stress in middle-aged women: a longitudinal study', Doctoral dissertation, Illinois Institute of Technology, Dissertation Abstracts International 38, 4537B. (University Microfilms No. 78--00, 865.).

Houston, J. P.: 1981, The Pursuit of Happiness (Scott, Foresman, Glenview, IL). Kammann, R.: 1982, 'Personal circumstances and life events as poor predictors of

happiness', Paper presented at the 90th annual convention of the American Psychological Association (August), Washington, D.C.

Kammann, R. and Fiett, R.: 1983, 'Affectometer 2: a scale to measure current level of general happiness', Australian Journal of Psychology 35, pp. 257--65.

Kanuk, L. and Berenson, C.: 1975, 'Mail surveys and response rates: a literature review', Journal of Marketing Research 12, pp. 440--53.

Kimmel, D. C., Price, K. F., and Walker, J. W.: 1978, 'Retirement choice and retirement satisfaction', Journal of Gerontology 33, pp. 575--585.

Kozma, A. and Stones, M. J.: 1980, 'The measurement of happiness: development of the Memorial University of New Foundland Scale of Happiness (MUNSH)', Journal of Gerontology 35, pp. 906--12.

Kozma, A. and Stones, M. J.: 1978, 'Some research issues and findings in the study of psychological well-being in the aged', Canadian Psychological Review 18, pp. 241-- 9.

Larsen, R. J.: 1983, 'Manual for the Affect Intensity Measure', Unpublished manuscript. Larsen, R. J., Diener, E., and Emmons, R. A.: 1983: 'An evaluation of subjective well-

being measures', Working Paper, University of Illinois at Champaign-Urbana. Larson, R.: 1978, 'Thirty years of research on the subjective well-being of older

Americans', Journal of Gerontology 33, pp. 109--25. Lawton, M. P.: 1975, 'The Philadelphia Geriatric Center Morale Scale: A revision',

Journal of Gerontology 30, pp. 85--9. Maddox, G. L. and Douglass, E. B.: 1973, 'Self-assessment of health: a longitudinal

study of elderly subjects', Journal of Health and Social Behavior 14, pp. 87--92. Mancini, J. A., and Orthner, D. K.: 1980, 'Situational influences on leisure satisfaction

and morale in old age', Journal of the American Geriatrics Society 28, pp. 466--71. Markides, K. S. and Martin, H. W.: 1979, 'A casual model of life satisfaction among the

elderly', Journal of Gerontology 34, pp. 86--93. McGill, V. J.: 1967, The Ideal of Happiness (Praeger, New York). McClure, R. F. and Loden, N.: 1982, 'Religious activity denomination membership and

life satisfaction', Psychology, A Quarterly Journal of Human Behavior 19, pp. 12-- 7.

McNeil, J. K., Stones, M. J., and Kozma, A.: 1986, 'Subjective well-being in later life: issues concerning measurement and prediction', Social Indicators Research 18, pp. 35--70.

Michalos, A. C.: 1980, 'Satisfaction and happiness', Social Indicators Research 8, pp. 385--422.

Michalos, A. C.: 1985, 'Multiple discrepancies theory', Social Indicators Research 16, pp. 347--413.

Michalos, A. C.: 1986a, 'Job Satisfaction, Marital Satisfaction, and the Quality of Life: A Review and Preview', in F. M. Andrews (Ed.), Research on the Quality of Life (University of Michigan Institute for Social Research: Ann Arbor), pp. 57--88.

Michalos, A. C.: 1986b, 'What makes people happy?' Paper presented at the Confer- ence on Quality of Life (October), Oslo, Norway.

LIFE S A T I S F A C T I O N M E A S U R E 57

Miller, J.: 1977, 'Studying satisfaction: modifying models, eliciting expectations, posing problems, and making meaningful measurements', in H. K. Hunt, The conceptualiza- tion of consumer satisfaction and dissatisfaction (Cambridge, MA: Marketing Sciences Institute).

Morgan, M.: 1984, 'Heavy television viewing and perceived quality of life', Journalism Quarterly 61, pp. 499--504, 740.

Morris, J. N. and Sherwood, S.: 1975, 'A retesting and modification of the Philadelphia Geriatric Center Morale Scale', Journal of Gerontology 30, pp. 77--84.

Near, J. P., Rice, R. W. and Hunt, R. G.: 1978, 'Work and extra-work correlates of life and job satisfaction', Academy of Management Journal 21, pp. 248--264.

Near, J. P., Rice, R. W., and Hunt, R. G.: 1980, 'The relationship between work and nonwork domains: a review of empirical research', Academy of Management Review 5, pp. 415--29.

Neugarten, B. L., Havighurst, R. J., and Tobin, S. S.: 1961, 'The measurement of life satisfaction', Journal of Gerontology 16, pp. 134--43.

Okun, M. A., Stock, W. A., Haring, M. J., and Witter, R. A. (in press), 'A quantitative synthesis of the social activity/subjective well-being relationship', Research on Aging.

Olsen, J. K.: 1980, 'The effect of change in activity in voluntary associations on life satisfaction among people 60 and over who have been active through time', Doctoral dissertation, University of Maryland, Dissertation Abstracts International 40, 5211A. (University Microfilms No. 80-07, 107).

Palmore, E.: 1979, 'Predictors of successful aging', The Gerontologist 19, pp. 427--31. Palmore, E. and Kivitt, V.: 1977, 'Change in life satisfaction: a longitudinal study of

persons aged 46--70', Journal of Gerontology 32, pp. 311--6. Palmore, E. and Luikart, C.: 1972, 'Health and social factors related to life satisfaction',

Journal of Health and Social Behavior 13, pp. 68--80. Parducci, A.: 1968, 'The relativism of absolute judgments', Scientific American 219, pp.

84--90. Parducci, A.: 1982, 'Toward a relational theory of happiness', Paper presented at the

90th Annual Convention of the American Psychological Association (August), Washington, DC.

Peters, G. R.: 1971, 'Self-conceptions of the aged, age identification and aging', The Gerontologist 11, pp. 69--73.

Peterson, J. L.: 1975, 'Personality effects self-esteem, need motivation, and locus of control on the life satisfaction of older black adults', Doctoral dissertation, Univers- ity of Michigan, Dissertation Abstracts International 35, 5700A. (University Microfilms No. 75--10,256).

Pomerantz, S. C.: 1978, 'Adolescent identity, self-esteem, and physical self-satisfaction as a function of age and sex: do they predict satisfaction with one's social milieu?', Doctoral dissertation, Temple University, Dissertation Abstracts International 39, 961B. (University Microfilms No. 78--12, 191).

Rahtz, D. R., Sirgy, M. J. and Meadow, H. L.: 1988, 'Elderly life satisfaction and television viewership: replication and extension', in S. Shapiro and H. H. Walle (eds.), 1988 AMA Winter Educator's Conference -- Marketing: A Return to the Broader Dimensions, (American Marketing Association, Chicago, IL), pp. 409--13.

Rahtz, D. R., Sirgy, M. J. and Meadow, H. L.: 1989, 'The elderly audience: correlates of television orientation', Journal of Advertising 18, pp. 9--20.

Ray, R. O.: 1979, 'Life satisfaction and activity/involvement: implications for leisure service', Journal of Leisure Research 11, pp. 112--9.

Reid, D. W. and Ziegler, M.: 1980, 'Validity and stability of a new desired control


measure pertaining to psychological adjustment of the elderly', Journal of Gerontol- ogy 35, pp. 395--402.

Rice, R. W., Near, J. P., and Hunt, R. G.: 1979, 'Unique variance in job and life satisfaction associated with work-related and extra-workplace variables', Human Relations 32, pp. 605--23.

Rice, R. W., Near, J. P., and Hunt, R. G.: 1980, 'The job satisfaction/life satisfaction relationship: a review of empirical research', Basic and Applied Social Psychology 1, pp. 37--64.

Richins, M. L.: 1987, 'Media, materialism, and human happiness', in S. Shapiro and H. H. Walle (eds.), 1988 AMA Winter Educator's Conference -- Marketing: A Return to the Broader Dimensions, (American Marketing Association, Chicago, IL), pp. 352--6.

Riddick, C. C.: 1980, 'The life satisfaction of retired and employed older women: a re- examination of the disengagement theory', Doctoral dissertation, Pennsylvania State University, Dissertation Abstracts International 41, 2327A. (University Microfilms No. 80-24,483).

Sauer, W.: 1977, 'Morale of the urban aged: a regression analysis by race', Journal of Gerontology 32, pp. 600--8.

Schwartz, N. and Clore, G. L.: 1983, 'Mood, misattributiom and judgments of well- being: informative and directive functions of affective states', Journal of Personality and Social Psychology 45, pp. 513--23.

Sirgy, M. J.: 1980, 'Toward a psychological model of consumer satisfaction/dissatisfaction', in R. Day and H. K. Hunt (eds.), New Findings on Consumer Satisfaction/Dis- satisfaction and Complaining Behavior, (Division of Research, School of Business, Indiana University, Bloomington, IN), pp. 40--7.

Sirgy, M. J.: 1983, Social Cognition and Consumer Behavior (Praeger, New York). Sirgy, M. J.: 1984a, Marketing as a Social Behavior: A General Systems Theory

(Praeger, New York). Sirgy, M. J.: 1984b, 'A Social Cognition Model of Consumer Satisfaction/Dissatisfac-

tion', Psychology and Marketing 1, pp. 27--44. Sirgy, M. J.: 1985, 'Using Self-Congruity and Ideal Congruity to Predict Purchase

Behavior', Journal of Business Research 13, pp. 196--206. Sirgy, M. J.: 1986, Self-Congruity: Toward a Theory of Personality and Cybernetics

(Praeger, New York). Sirgy, M. J.: 1987a, 'Toward a general systems theory of social behavior: A psycho-

cybernetic perspective', Systems Research 4, pp. 93--110. Sirgy, M. J.: 1987b, 'A social cognition model of consumer problem recognition', The

Journal of the Academy of Marketing Science 15, pp. 53--61. Sirgy, M. J. and Tyagi, P.: 1986, 'An attempt toward an integrated theory of consumer

psychology and decision-making', Systems Research 3, pp. 161--75. Spreitzer, E. and Snyder, E. E.: 1974, 'Correlates of life satisfaction among the aged',

Journal of Gerontology 29, pp. 454--8. Stock, W. A., Okun, M. A., Haring, M. J., and Witter, R. A.: 1983, 'Age and subjective

well-being: a meta-analysis', in R. J. Eight, ed., Evaluation Studies: Review Annual, (Sage, pp. 279--302).

Tellegen, A.: 1979, 'Differential Personality Questionnaire', Unpublished manuscript. Toseland, R. and Rasch, J.: 1979--1980, 'Correlates of life satifaction: an AID analysis',

International Journal of Aging and Human Development 10, pp. 203--11. Underwood, B. and Froming, W. J.: 1980, 'The Mood Survey: a personality measure of

happy and sad moods', Journal of Personality Assessment 44, pp. 404-- 14. VanCoevering, V. G. R.: 1974, 'An exploratory study of middle-aged and older widows

to investigate those variables that differentiate high and low life satisfaction',


Doctoral dissertation, Wayne State University, Dissertation Abstracts International 34, 3895A. (University Microfilms No. 73--31,788).

Weaver, C. N.: 1978, 'Job satisfaction as a component of happiness among males and females', Personnel Psychology 31, pp. 831--40.

Wessman, A. E.: 1957, 'A psychological inquiry into satisfaction and happiness', Doctoral dissertation, Princeton University, Dissertation Abstracts International 17, 1-384. (University Microfilms No. 00--20, 168).

Wicklund, R. A. and Brehm, J. W.: 1976, Perspectives on Cognitive Dissonance (Lawrence Erlbaum Associates, Hillsdale, NJ).

Wilson, W.: 1967, 'Correlates of avowed happiness', Psychological Bulletin 67, pp. 294--306.

Wilson, W. R.: 1960, 'An attempt to determine some correlates and dimensions of hedonic tone', Doctoral dissertation, Northwestern University, Dissertation Abstracts 22, 2814. (University Microfilms No. 60-6588).

Wortman, C. and Silver, R.: 1982, 'Coping with undesirable life events', Paper presented at the 90th Annual Convention of the American Psychological Association (August), Washington, DC.

Wyer, R. S.: 1975, Cognitive Organization and Change: An Information Processing Approach (Lawrence Erlbaum Associates, Potomac, MD).

Zeglan, M. E.: 1977, 'The impact of primary relationships on life satisfaction of the elderly', Doctoral dissertation, Washington State University, Dissertation Abstracts International 37, 5372A. (University Microfilms No. 77-2892).

Virginia Polytechnic Inst., State University, Center for Strategy &Marketing Studies, Department of Marketing, Blacksburg, VA 24061-0236 U.S.A.

Documents

A life satisfaction measure based on judgment theory