PURPOSE IN LIFE TEST

Variable:

The Purpose in Life Test (PIL) was developed to assess the degree to which an individual possesses meaning, under­ stood as the opposite of existential frustra­tion or a lack of fulfillment of the will to meaning (e.g., Frankl, 1955, 1960). Higher scores on the PIL are taken to indicate less of a presence of existential vacuum, which is a state of emptiness, manifested chiefly by boredom (Crumbaugh & Henrion, 1988). Frankl viewed having meaning as a funda­mental human motivation. Possession of a substantial degree of this is seen as a typical condition for normally functioning individ­uals, and one that may be absent in clinical populations. Note that the PIL is designed to measure the extent to which meaning has been found, not the motivation to find such purpose.

A theoretical connection between reli­gion and purpose in life is made by Frankl (1955, 1958), as he indicates that this need can best be understood as spiritual, although it is not clear how the term “spiritual” is used. This association is empirically sup­ ported with findings reported by Crandall and Rasmussen (1975). Respondents who scored high on the PIL indicated that salva­tion was a relatively higher value on the Rokeach (1967) Value Survey than did individuals scoring lower on the PIL. Further, they found the PIL to be significantly corre­lated with having a more intrinsic orienta­tion to religion, but independent of an ex­trinsic orientation (see, e.g., Allport & Ross, 1967, for discussion of these religious ori­entations).

The PIL has been the primary self-ad­ ministered assessment of purpose and meaning in life (Chamberlain & Zika, 1988; Crumbaugh & Henrion, 1988; Dyck, 1987). New measures are being developed, but none yet has the empirical foundation of the PIL. In their evaluation of measures of meaning, Chamberlain and Zika (1987) con­cluded that the best general measure was the PIL but that there are a number of compo­nents to meaning in life that should be more carefully evaluated. This is consistent with the view expressed by Crumbaugh and Henrion (1988) that as new measures are devel­oped, it may become clear which tool is best suited for various purposes.

Description:

The PIL has three parts. The objectively scored Part A (which is the part most frequently used for research and is the focus of this review) is composed of 20 items. Respondents indicate on seven point scales how much they experience the con­ tent described by the item. Each item has its own anchor points. For example, one item begins with the stem, “I am usually:” and subjects respond on a 1- 7 scale, anchored by “completely bored” for 1 and “exuber­ant, enthusiastic” for 7. Four is labeled “neutral” on all scales. These responses are summed to determine the overall score.

Part B is 13 incomplete sentences to which the respondent provides completion. Part C has the individual write about his or her “life goals, ambitions, hopes, future plans, what has provided them meaning in the past, and what could motivate them in the future” (Crumbaugh & Henrion, 1988, p. 78).

Practical Considerations:

This self-admin­istered questionnaire requires a fifth-grade reading level ability. No time limit is typi­cally given, but most respondents finish in less than 15 minutes (Crumbaugh & Hen­rion, 1988).

Because the Part A format has each item utilize different end points for the scale, it may be more confusing to respondents than a measure that repeatedly uses the same end points; this format also makes the test more awkward and bulky when used in group ad­ ministrations (Harlow, Newcomb, & Bent­ler, 1987).

Parts B and C must be evaluated by a clinician and are ignored for most research purposes. Attempts to quantify these latter sections have added little to what is gained by part A (Crumbaugh & Henrion, 1988).

Norms/Standardizotion:

The original study used five groups assumed to vary in life purpose to establish norms on the PIL (Crumbaugh & Maholick, 1964). Specifi­cally, running from “high purpose” on down were (a) Junior League women and men from Frankl’s Harvard seminar, (b) Harvard undergraduates, (c) outpatients of various psychologists and psychiatrists in Georgia, outpatients of a nonprofit psychiatric clinic in Georgia, and (e) hospitalized alco­holic patients. Using these groups, Crum­ baugh and Maholick (1964) reported means of 119 for nonpatients and 99 for patients. Combining these people with a large num­ber of others, Crumbaugh (1968) reported the mean for normals as 112.4 (SD= 14.l) and for patients as 92.6 (SD = 21.3). Moti­vated business and professional people (M = 118.9) and trainees for a religious order (M = 119.3) score high on the PIL (Crumbaugh, Raphael, & Shrader, 1970). Prison inmates have twice been found to score, around 100 (Reker, 1977). Crum­baugh and Henrion (1988) give a “cutting score” of 102 between clinical and normal populations. The estimated standard devia­tion noted is 19. Scores at or above 113 in­dicate a definite presence of purpose and meaning, and scores at or below 91 indicate a lack of clear meaning or purpose.

The age of respondent should be consid­ered when interpreting results. Adolescent scores (ages 13-19, M = 104.l) have been found to be lower than scores of older groups (25 and older, M = 111.5; Meier & Edwards, 1974), although associations with age are not always found (Crumbaugh & Henrion, 1988). Seldom have gender differ­ences been statistically significant (Crum­ baugh & Henrion, 1988; Meier & Edwards, 1974). Educational level is not clearly re­ lated to PIL scores.

Reliability:

The original split-half reliability reported using the odd-even method was .81, Spearman-Brown corrected to .90 (Crumbaugh & Maholick, 1964). In a sample of inmates, Reker (1977) reported a solid split-half reliability of .85, corrected to .92; this was also found by Crumbaugh (1968) in a sample of 120 “active and lead­ing Protestant parishioners” (p. 75). Test­ retest correlations have been .68 (12 week; Reker, 1977), .79 (6 week; Reker & Cousins, 1979) and .83 ( I week; Meier & Edwards, 1974).

Validity:

Though there is room for ques­tions, the PIL appears to be a reasonably valid measure. That the PIL assesses some­ thing of psychological significance is sup­ ported by its ability to predict membership in clinical versus nonclinical populations. For example, it distinguishes between psy­chological patients and nonpatients (Crum­ baugh, 1968; Crumbaugh & Maholick, 1964), and inmates and noninmates (Reker, 1977). It is also correlated with therapists’ (r = .38) and clergy’s (r = .47) ratings of indi­viduals’ possession of meaning or purpose in life (Crumbaugh, 1968). Further, a large number of studies have found it to be mod­erately associated with variables that the lo­gotherapy suggests are tied to meaning.

The PIL is not identical to other con­ structs, as indicated by its low association with other measures. For example, Crum­baugh and Maholick (1964) report that scores on the PIL are not highly related to MMPI scores. It is also only moderately re­lated to the desire to find meaning (Crum­ baugh, 1977). In the original study, the high­ est association was with depression (r = .39); a follow-up study with normal and psychi­atric patients found a higher association with depression (r = .65; Crumbaugh, 1968). The correlation between the PIL and depression is so consistent that Dyck (1987) views the PIL as primarily an indirect measure of depres­ sion. This connection may be what is respon­sible for its relation to clinical status.

There is some debate about the degree to which the PIL is contaminated by a socially desirable response bias. Crumbaugh and Henrion (1988) report several studies show­ing mixed results. As the PIL is self-admin­istered, there is certainly the ability for indi­viduals to manipulate their answers. They point out that more competitive situations may be more likely to bring out this associ­ation.

When the PIL is factor analyzed, it ap­ pears to be made up of a number of discern­ able components (e.g., Dyck, 1987; Reker & Cousins, 1979). This suggests a lack of conceptual coherence in the tool (Dyck, 1987). If used, it should be kept in mind that only a portion of the measure may be di­rectly assessing purpose in life and that other components (e.g., depression) may be influencing the degree of association of the PIL to other variables of interest.

The PIL was developed using primarily white American respondents, and it is based on a culturally Western philosophical view. Thus there are questions about its validity among other populations. Researchers have begun the task of translating and modifying the instrument for such work, and the de­gree to which this can be successful likely will be known soon. This concern should be considered when using the specified cut-off scores. As one’s sample diverges from the population used to create those such scores, the meaning of a particular score becomes more ambiguous.

In sum, the PIL is empirically associated with theoretically linked variables, and it is not identical to what is measured by other means. However, a more complete evalua­tion of the coherence and validity of the test is needed.

Location: The PIL was originally described in Crumbaugh and Maholick (1964). The PIL and manual are published by Psycho­ metric Affiliates [P.O. Box 807, Murfrees­boro, TN 37133), and can be ordered from them, or the Institute of Logotherapy in Berkeley, California. The version below was taken from Nackord and Fabry (1983).

Subsequent Research:

References