SPSS Assignment Part 2 Instructions
Evaluating Reliability and Validity
When tests are created they usually are evaluated for their usefulness. To determine the usefulness of a test the authors and/or publishers typically assess itspsychometric properties. Psychometric properties refer to quantifiable characteristics of a test that relate to its statistical strength or weakness. The main two features that are used to determine whether or not a test is statistically strong are reliability and validity. As you recall from your textbook readings there are various types of reliability and validity as well as various ways to quantify each.
For the present assignment you will be evaluating the reliability and validity of the CES-D using the fictitious data set provided in Bb. In order to assess/quantify the reliability of the CES-D you will be calculating a statistic called Cronbachs Alpha. Cronbachs Alpha is used to estimate a type of reliability called internal consistency or the degree to which each test question consistently measures the same underlying construct(s). In order to assess/quantify the validity of the test you will be calculating a statistic called Pearsons Correlation Coefficient between total scores on the CES-D and scores on the Beck Depression Inventory (BDI) another test designed to measure depressive symptoms. When researchers find the correlation between two tests designed to measure the same underlying construct(s) they are providing evidence for the convergent validity of the tests.
Before beginning your analyses please review the 20 questions that comprise the CES-D Scale. When we sum participants scores on all 20 questions we obtain a Total Score. Several researchers agree that in addition to a total score participantsscores can be categorized into four (4) subscales: Depressive Affect (DA) Well-Being (W) Somatic (S) and Interpersonal (I) (e.g. Hertzog Van Alstine Usala Hultsch and Dixon 1990; Knight Williams McGee and Olaman 1997). Each subscale comprises a different subset of questions and assesses a different feature of depressive symptomology. The table below indicates which questions on the CES-D Scale correspond to each subscale.
CES-D Subscales
DA = Depressive Affect W = Well-being S = Somatic I = Interpersonal
Q#1 S
Q#2 S
Q#3 DA
Q#4 W
Q#5 S
Q#6 DA
Q#7 S
Q#8 W
Q#9 DA
Q#10 DA
Q#11 S
Q#12 W
Q#13 S
Q#14 DA
Q#15 I
Q#16 W
Q#17 DA
Q#18 DA
Q#19 I
Q#20 S
For this assignment you will:
Be sure to carefully review the Grading Rubric posted in the Assignment Instructions folder so that you know exactly what is expected of you when completing this assignment.
This assignment is due by 11:59 p.m. (ET) on Monday of Module/Week 3.
References
Hertzog C.; Van Alstine J.; Usala P.D.; Hultsch D.F.; &Dixon R. (1990).Measurement properties of the Center for Epidemiological Studies Depression Scale (CES-D) in older populations.Psychological Assessment: A Journal of Consulting and Clinical Psychology 2(1) 64-72. doi: 10.1037/1040-3590.2.1.64
Knight R. G. Williams S. McGee R. & Olaman S. (1997). Psychometric properties of the
Centre for Epidemiologic Studies Depression Scale (CES-D) in a sample of women in
middle life.Behavior Research and Therapy 35(4) 373380.doi:10.1016/S0005-
7967(96)00107-6