Construct Validity Assignment
Chapter 5 introduced several concepts that are important for establishing reliable and valid
measurement. In this assignment you will apply what you have learned to a personality measure
called the Ten-Item Personality Inventory (TIPI).
Type your answers in a new document. Be sure that your work is neatly organized with each
question number clearly labeled. Save the file as a Microsoft Word Document or a PDF, and attach
the file to the BbLearn dropbox. Other file formats and email submissions will not be accepted.
First, complete an online version of the TIPI at the following URL. Your results will include an
overview of the “big five” personality traits. http://psychcentral.com/quizzes/personality.htm
_____ 1. Extraverted, enthusiastic
_____ 2. Critical, quarrelsome
_____ 3. Dependable, self-disciplined
_____ 4. Anxious, easily upset
_____ 5. Open to new experiences, complex
_____ 6. Reserved, quiet
_____ 7. Sympathetic, warm
_____ 8. Disorganized, careless
_____ 9. Calm, emotionally stable
_____ 10.Conventional, uncreative
1. The TIPI measures each trait with two items, one of which is reverse-worded. For example,
items #1 and #6 measure extraversion. Higher scores on item #1 reflect greater extraversion.
Item #6 is reverse-worded so that lower scores reflect greater extraversion.
a. Items #4 and #9 measure emotional stability. If higher scores should reflect greater
emotional stability, which of these two items is reverse-worded?
b. Which items appear to measure openness to experience? Which of these seems to be
reverse-worded?
Questions #2-4 refer to the following passage: Gosling, Rentfrow, and Swann (2003) developed the
TIPI with the goal of creating a brief personality measure. Some previously developed measures are
considerably longer. For example, a personality measure called the NEO-PI-R contains 240 items
and takes about 45 minutes to complete. The NEO-PI-R is a well-established measure with good
construct validity, but it is too lengthy for some research studies. Gosling et al. (2003) state, “In an
ideal world, personality researchers would have sufficient time and resources to exploit the
superior content validity and reliability of well-established multi-item instruments. Unfortunately,
circumstances are often not ideal and researchers may be faced with a stark choice of using an
extremely briefly instrument or using no instrument at all.” When developing the TIPI, Gosling et al.
(2003) carefully evaluated the measure’s reliability and validity. The following questions describe
some of that research.
2. Gosling et al. (2003) report that the Cronbach’s alpha for the emotional stability scale of the TIPI
was 0.73. What does this tell us about the reliability of the TIPI? Be sure to indicate which
specific aspect of reliability is assessed by this value and explain what this result means.
I see myself as:
Response scale:
1 (disagree strongly) to 7 (agree strongly)
3. When evaluating the construct validity of the TIPI, the researchers administered the TIPI to a
sample of University of Texas at Austin students. The students completed the measure on two
occasions, with the second measure obtained six weeks after the first test administration. The
researchers then correlated the participants’ personality scores from the first session with their
scores from the second session.
a. Which specific aspect of measurement reliability is being described here?
b. The average correlation was 0.72. What does that tell us about the TIPI’s reliability?
4. Gosling et al. (2003) also examined the relationship between trait scores on the TIPI and trait
scores on another well-established, validated, and longer personality measure called the Big
Five Inventory (BFI). The results of these analyses are presented in the table below.
a. Examining the relationship between extraversion on the TIPI and extraversion on the BFI
(r = .87) allows us to evaluate which specific aspect of the TIPI’s construct validity?
b. Examining the relationship between openness to experience on the TIPI and extraversion
on the BFI (r = .36) allows us to evaluate which specific aspect of the TIPI’s construct
validity?
c. Which of the five traits has the strongest convergent validity? Explain what this means.
d. Which TIPI trait is most clearly discriminant from the Big-Five extraversion trait? (Hint:
use only the correlations in the first column, and think carefully about the definition of
discriminant validity.)
5. Suppose that we examine whether the trait scores on the TIPI are related to several behavior
variables that we measure at a later date. For example, we could test whether students’
extraversion scores relate to their rates of participation in campus activities. We could also test
whether conscientiousness scores upon entering college relate to GPA at graduation. Which
specific aspect of construct validity would we be evaluating by testing for those relationships?
6. Some research examining the personality traits of leaders has shown that leaders tend to show
relatively high conscientiousness. Suppose that we want to apply this finding to help us
evaluate the construct validity of the TIPI using the known-groups paradigm. Specifically, we
recruit a sample of student leaders and non-leaders and compare their scores on the
conscientiousness subscale. Which of the following results would provide the strongest validity
evidence?
a. Leaders (M = 5.80, SD = 0.42) & Non-leaders (M = 5.70, SD = 0.55), p < .05
b. Leaders (M = 5.80, SD = 0.42) & Non-leaders (M = 5.70, SD = 0.55), p > .05
c. Leaders (M = 5.80, SD = 0.42) & Non-leaders (M = 4.50, SD = 0.55), p < .05
d. Leaders (M = 5.80, SD = 0.42) & Non-leaders (M = 4.50, SD = 0.55), p > .05
7. The result described in question 6 would provide evidence for the measure’s ____________
validity.