Question Type 9: Determining whether certain issues pertain to validity or reliability Exercises

A weighing scale adds exactly 10 kg to every measurement. Determine whether the measurements are reliable, valid, both, or neither, and justify your answer.

Two examiners independently grade 20 essays on a 0–10 scale and their scores are highly consistent. Identify the type of reliability and explain your answer.

A final exam omits half of the course topics. Identify which type of validity is compromised and explain your reasoning.

Students who did well on yesterday’s test perform similarly on today’s equivalent test. Identify the type of reliability this scenario illustrates and explain why.

A cohort of students with high SAT scores earn low GPAs in their first university year. Identify the type of validity concern and justify your answer.

A new happiness scale shows very low correlation with an established depression scale. Name the validity type illustrated and justify.

A new anxiety inventory correlates strongly with an established anxiety measure. Which type of validity does this demonstrate and why?

Five subjects’ weights were measured twice by the same scale. The paired measurements (in kg) are: Subject A: (60, 61) Subject B: (70, 71) Subject C: (80, 79) Subject D: (90, 91) Subject E: (50, 49)

Compute the test–retest reliability coefficient using Pearson’s $r$ and interpret the result.

Six students’ SAT scores (out of 1600) and their first‐year GPAs (out of 4.0) are: (1400, 3.5), (1200, 3.0), (1500, 3.8), (1100, 2.8), (1300, 3.2), (1000, 2.5).

Compute the Pearson correlation coefficient and comment on the predictive validity of SAT scores for first‐year GPA.

Five students took two parallel forms of a test. Their scores out of 100 are: Student 1: (90, 88), Student 2: (75, 78), Student 3: (82, 80), Student 4: (68, 70), Student 5: (95, 92).

Compute the parallel‐forms reliability coefficient (Pearson’s $r$) and state its implication.

A test consists of three items administered to five students. Their item scores are:

Compute Cronbach’s alpha for internal consistency and interpret.

Two raters independently classify 10 essays as Pass (P) or Fail (F). Their decisions form the table:

Compute Cohen’s kappa and interpret the inter‐rater reliability.