Adlerian Personality - T.O.P PDF

Preview

Summary

T.O.P...

Description

Exploring the Test–Retest Reliability of the Adlerian Personality Priority Assessment

Maggie Parker and A. Elizabeth Crunk

Abstract The authors assessed the test–retest reliability of the Adlerian Personality Priority Assessment (APPA) with a convenience sample of university students (N = 120). Participants completed the APPA twice within a 3-week interval. Intraclass correlation (two-way mixed model) was used to measure test–retest reliability. The APPA demonstrated good temporal reliability for all four factors, as well as initial evidence of convergent validity. The results provide further evidence for counselors’ use of the APPA to identify individuals’ personality priorities. Keywords: Adlerian Personality Priority Assessment, test–retest reliability, intraclass correlation, convergent validity, Adlerian theory, personality assessment

Personality priority assessments can be used to better understand the ways individuals view themselves, others, and the world. Dillman Taylor, Ray, and Henson (2015) developed the Adlerian Personality Priority Assessment (APPA) to assess individuals’ personality priorities consistently with Adlerian theory. The APPA improved on several measurement-related limitations of previous personality priority assessments; however, test–retest reliability of the scale has not yet been established. The purpose of this article is to examine the test–retest reliability of the Adlerian Personality Priority Assessment with a sample of undergraduate and graduate students.

Personality Priorities According to Adler’s theory of Individual Psychology (1931), individuals have an innate drive to belong, feel worthy, and strive for significance. As children develop through social interactions, they form perceptions of how best to achieve that significance and belonging (Kottman & Ashby, 1999). They begin to formulate a style of life, or a set of beliefs about themselves, others, and the world (Kottman, 2003). This style of life helps individuals understand and interpret events and experiences throughout their lives (Mosak & Maniacci, 2010; Watts, 1999).

The Journal of Individual Psychology, Vol. 75, No. 3, Fall 2019 © 2019 by the University of Texas Press Editorial office located in the College of Arts and Sciences at Lynn University. Published for the North American Society of Adlerian Psychology.

232 Maggie Parker and A. Elizabeth Crunk

A goal within Adlerian therapy is for counselors to help clients gain insight into themselves and their style of life to confront and hopefully change the mistaken beliefs that hinder their self-enhancement (Kottman, 2003). To understand clients more fully, and to gain insight into clients’ style of life, Adlerian theorists utilize a variety of informal and formal assessments. These include an exploration of family dynamics, early recollections, an exploration of mistaken beliefs, social interest, and typologies (Ansbacher & Ansbacher, 1956; Mosak & Maniacci, 2010). Adlerian counselors can use typologies, also known as personality priorities, to conceptualize their clients, and individuals can use them to better understand themselves (Kefir, 1971; Pew, 1976). According to Adlerian theorists, individuals identify with at least one of the following four personality priorities: controllers, avoiders, superiors, and pleasers (Kefir, 1971; Pew, 1976). These four priorities provide insights into ways individuals interact with others and the world around them as they strive for significance. Pew (1976) expanded on these priorities, labeling them “control,” “superiority,” “pleasing,” and “comfort.” Dillman Taylor (2013) provided descriptions of each priority and how individuals within each priority tend to strive for significance. Individuals with a priority of control enjoy leadership positions, tend to take charge of situations, and prefer positions of authority (Dillman Taylor, 2013). Those with pleasing as their primary priority often strive to meet others’ needs. Their goals are often to ensure that others are pleased, and they often experience anxiety if others do not like them. Those who identify as superiority attempt to accomplish their goals through achievements and the completion of self-identified tasks, and reaching selfidentified goals provides meaning in their lives. Individuals with the comfort priority prefer lower-stress jobs, activities, and projects. They tend to avoid activities that will place demands on them (Dillman Taylor, 2013).

Adlerian Personality Priority Assessment The APPA (Dillman Taylor et al., 2015) measures personality priorities consistent with Adler’s (1931) Individual Psychology theory of personality, and it was developed to address measurement-related limitations of prior personality priority assessments. The Langenfeld Inventory of Personality Priorities, developed by Langenfeld and Main (1983), was the first formalized assessment to provide empirical support for personality priorities. Despite empirical support for the construct of personality priorities, less than a quarter (22%) of the variance accounted for was explained by the items on the factors (Langenfeld & Main, 1983), thus failing to meet the criteria for a valid and reliable measure of personality (Dimitrov, 2012). In response, Dillman Taylor et al. (2015) created the APPA. The APPA is a 30-item instrument that

Test–Retest Reliability of the APPA 233

assesses Adlerian personality constructs across the four Adlerian personality priority domains: pleasing, comfort, control, and superiority (Table 1). Dillman Taylor et al. (2015) conducted an exploratory factor analysis of the APPA with a sample of undergraduate students (N = 393), finding initial empirical support for the four-factor structure of personality priorities most commonly used in the literature: superiority, control, comfort, and pleasing. Cross-validation of the exploratory model using confirmatory factor analysis (CFA; Dillman Taylor, 2013) indicated good model fit across four fit indexes (Hu & Bentler, 1999), which provided further support for the four-factor structure. Dillman Taylor’s (2013) study also provided initial evidence of discriminant validity as well as of the APPA’s generalizability across participant demographics in that sample. Dillman Taylor (2013) asserted the need for additional tests of reliability, including test–retest. Although the APPA has demonstrated satisfactory psychometric properties with acceptable internal consistency reliability with graduate (Dillman Taylor, Gungor, Blount, & Mullen, 2018) and undergraduate students (Dillman Taylor, 2013; Dillman Taylor et al., 2015), its longitudinal stability has not been examined. Evidence of test–retest reliability is needed to demonstrate whether the APPA is a consistent measure of static personality priorities across time. Therefore, in the present study, we examined the test–retest reliability of the APPA, administering the APPA twice over 3 weeks with a sample of undergraduate and graduate students.

Purpose of the Study Building on previously reported psychometric properties of the APPA, the present study examined the test–retest reliability of the APPA with a sample of undergraduate and graduate students using intraclass correlation (ICC; Koo & Li, 2016) procedure. Schmidt, Le, and Ilies (2003) recognized that directing attention to retest reliability of assessments can provide information on assessments’ ability to capture the stability of personality priority over time. It is important for researchers to assess meaningful, stable individual differences (Rae & Olson, 2018), thus making the establishment of adequate test–retest reliability of the APPA critical. To explore the temporal stability of a measure using the test–retest method, it is necessary to have a sample evaluated at least twice over a period when the construct evaluated remains stable (Watson, 2004). Watson (2004) noted that to assess test–retest reliability, the administration of tests must be within a close enough time frame that the traits being measured would not be expected to change. John and Robins (1993) noted that traits are more temporally reliable if they are central to the individual’s identity and more evaluatively neutral, such as those assessed within the APPA (Dillman

Table 1 Adlerian Personality Priorities Personality Priority

Definition

Pleasing

People with this priority are engaging, outgoing, and gregarious. They attempt to connect with others and help others feel comfortable. They tend to fear rejection of others and attempt to do as they believe others expect. At times, their attempts to make others happy result in feeling alienated, and their ideal self often is different from their self-perception. When non-pleasing individuals are around pleasers, they tend to feel exasperated after continued demands for approval. Assessment examples include “I work really hard to keep from offending others”; “ I have a strong desire to meet others’ needs”; “I need to know that others are pleased with me”

Control

Individuals with control personalities seek to control themselves, others, and/or situations. They are highly organized, assertive, and often seek out leadership roles. They may experience difficulty enjoying spontaneity and creativity. They seek to avoid surprises and often fear ridicule from others. Individuals often feel tense or resistant when around control personalities. Assessment examples include: “I tend to take charge”; “I can make things happen by taking charge”; “In most situations I prefer to be in charge”

Comfort

Individuals who identify with comfort personalities are often calm and easygoing. They are peaceful and often experienced by others as laidback. Comfort priorities avoid stress and do not like expectations, sometimes being unproductive in their efforts to avoid stress. Others often experience comfort priorities as frustrating because of their lack of follow-through, irritating, or boring. Assessment examples include: “I avoid projects that are stressful”; “I prefer not having a lot of work to do”; “If things are tough or difficult, I move on to something else”

Superiority

Individuals with superiority personalities strive to be the best in most, if not all, situations. They are knowledgeable and idealistic. They may feel overburdened by responsibilities, as they often take over tasks to complete them “the right way.” They want to avoid meaninglessness in life. When around superiority personalities, others may feel less than, or experience a sense of not being as good or good enough. Assessment examples include: “I need to be the winner in games”; “My goal in life is to be the best”; “I often feel my ideas are better than others”

Adapted from Dillman Taylor, Ray, and Henson (2015).

Test–Retest Reliability of the APPA 235

Taylor et al., 2015). With these considerations in mind, the researchers utilized a short time period between testing to reduce the potential impact of developmental or environmental impact on individuals’ personality trait consistency. We identified a 3-week period to allow for sufficient time between testing.

Method Participants Participants included 120 undergraduate and graduate students from a small, private, religiously affiliated university in the southeastern United States. Upon approval from the institutional review board, the first author sent out a schoolwide email inviting students to participate in a research project examining personality priorities. The initial sample size after the first administration was 325. After 3 weeks, another email was sent to the student body, asking for them to complete the assessment. Three hundred and thirtysix students completed the assessment at the second administration. Thus, a total of 661 students completed the APPA across the two assessment points. Of the 661 students that completed the APPA, 124 completed it twice, providing ample sample size for assessing the APPA’s test–retest reliability. Four cases contained incomplete data sets and were deleted before analysis, resulting in a sample of 120. The final sample of 120 participants exceeds the recommended sample size of 30 participants needed to conduct ICC (Koo & Li, 2016). Of the 120 participants in our final sample, most identified as White (n = 96, 80%) and female (n = 92, 76.7%). Participant year in school was split fairly evenly. Additional demographic information is presented in Table 2. Instrumentation After completing the demographics questionnaire, participants were instructed to complete the APPA and the Perceived Wellness Survey (PWS) (Adams, Bezner, & Steinhardt, 1997). Adlerian Personality Priority Assessment. The APPA is a 30-item self-report instrument that uses a 5-point Likert scale asking respondents to answer each item according to how true the item is for them, where 1 = not at all, 2 = a little bit, 3 = somewhat, 4 = quite a bit, and 5 = very much. Items on the APPA loaded onto four subscales aligned with Adlerian theory: pleasing, comfort, control, and superiority. Sample items include “I tend to take charge” (i.e., control) and, “I tend to avoid committing to too many things” (i.e., comfort). Scores were calculated by totaling each scale and taking the average; a higher score for a specific subscale indicates a greater level of that particular personality priority (Dillman Taylor, 2013; Dillman Taylor et al., 2015).

236 Maggie Parker and A. Elizabeth Crunk

Table 2 Participant Demographics Demographic

n

%

Gender

Male Female Other

Ethnicity

Caucasian African American Asian American Indigenous American Other Freshman Sophomore Junior Senior Graduate student

27 92 1 96 12 8 2 2 29 17 27 27 19

22.5 76.7 .8 80.0 10.0 6.7 1.7 1.7 24.2 14.2 22.5 22.5 15.8

18–20 21–23 Older than 24

55 42 23

45.8 34.9 19.3

Classification

Age

The APPA has demonstrated acceptable to good internal consistency reliability in a sample of undergraduate students (N = 393), with Cronbach’s alpha values ranging from .80 to .91 across all four subscales (Dillman Taylor et al., 2015). In a study examining the relationship between wellness and personality priority among counselors-in-training, internal consistency reliability of the APPA was found to be satisfactory to good, with Cronbach’s alpha levels ranging from .71 to .91 across all subscales (Dillman Taylor et al., 2018). Perceived Wellness Survey. Convergent validity of the APPA was assessed using the PWS (Adams et al., 1997), a 36-item self-report scale that assesses self-perceived well-being on six domains: physical, social, emotional, intellectual, psychological, and spiritual. Items are scored on a 6-point Likert scale ranging from 1 (very strongly disagree) to 6 (very strongly agree), with higher total and subscale scores suggesting greater wellness. The PWS has demonstrated good internal consistency reliability, with Cronbach’s alphas of .68 to .86 across subscales and .94 for the total scale. In the present study, the PWS demonstrated excellent internal consistency reliability for the total scale (α = .936) and acceptable to good internal consistency reliability

Test–Retest Reliability of the APPA 237

across subscales (with the exception of the intellectual wellness subscale, α = .68), including psychological wellness (α = .76), emotional wellness (α = .82), social wellness (α = .72), physical wellness (α = .84), and spiritual wellness (α = .86). Procedures The first author obtained permission from the developer of the APPA (Dillman Taylor, 2013) to examine the scale’s test–retest reliability. The APPA was then uploaded to an online platform to increase student access and study participation. The online instructions were provided exactly as they appear on the paper-and-pencil format. After receiving consent from the university’s institutional review board, the first author emailed the student body. Because of the online platform, students completed the survey at their leisure and the location of completion is unknown. Three weeks after the first email was sent, the first author sent a second email requesting that students complete the survey. To reduce potential for bias, feedback was not provided between administrations. Statistical Analysis Using SPSS (Version 21.0), we inspected the data and explored the nature of the variables to ensure that the data met the assumptions for the statistical analysis used to address the research questions. We examined standardized residual plots to assess for linearity and homoscedasticity. We inspected Q–Q plots and computed a Kolmogorov-Smirnov goodness-of-fit test to assess for normality. We examined bivariate correlations and variance inflation factors to assess for multicollinearity. The collective findings of these preliminary analyses indicated the model assumptions had not been violated. As a result, the data were deemed appropriate to analyze using ICC. ICC is a reliability index widely used in test–retest research (Koo & Li, 2016). Although researchers have often used Pearson correlation coefficient to explore test–retest reliability, it is not an ideal measure for reliability (Yen & Lo, 2002). ICC, however, can reflect both the degree of correlation and agreement between measurements, thus making it a more sensitive statistical procedure when exploring test–retest reliability (Koo & Li, 2016; Yen & Lo, 2002). When exploring test–retest reliability, researchers attempt to detect the effect of error rather than correlation between two measures; thus, ICC is a more appropriate statistical analysis for addressing the test–retest reliability of the APPA (Yen & Lo, 2002). Because repeated measurement is not equated to randomized samples, a two-way mixed model is utilized to explore the test–retest reliability of the APPA (Koo & Li, 2016; Portney & Watkins, 2000). In the present study, we analyzed data using a single-measurement, absolute-agreement, two-way mixed-effects model. Because the APPA is a four-factor model, individual

238 Maggie Parker and A. Elizabeth Crunk

ICC tests were computed for each of the four factors of the APPA across two test administrations.

Results Internal Consistency Reliability In this study, the Cronbach’s alpha coefficients remained stable at both testing times. The coefficient was .82 for the total scale at both testing times, and .74 for comfort wellness items at both testing times. For superiority items, Cronbach’s alpha was .76 at Time 1 and .80 at Time 2. For control items, α = .88 at Time 1 and .89 at Time 2, and within pleasing items, α = .90 at Time 1 and α = .92 at Time 2. These results indicate good total score internal consistency and acceptable internal consistency across the subscales and testing times. Test–Retest Reliability Table 3 presents test–retest reliability values of the four personality scales for the two administrations of the APPA. For the total sample, ICCs indicated good reliability for all four factors within the APPA, ranging from .65 to .72 (Koo & Li, 2016). The personality priority of pleasing resulted in the highest test–retest reliability. Within pleasing, the intraclass correlation for single measures was .72. Within the 95% confidence interval, the lower bound score was .63 and the upper bound was .80, indicating good reliability. Control and comfort had the lowest

Table 3 Results of ICC Calculation in SPSS Using Average-Rating Linear Relationship, Two-Way Random-Effects Model

Intraclass

Correlation

Single Measures

95% Confidence Interval Lower Bound

Upper Bound

F-Test With True Value 0 Value

df1

df 2

Sig.

Pleasing

.72

.63

.80

6.23

119

119

.000

Control

.65

.53

....