F23 Ch 2 and 3: Variability and Measurement

In an experiment, which of the following does not contribute to error variance?

• The personalities of the participants
• The variable of interest
• The weather
• Random equipment malfunctions

In an experiment on aggression, the researcher was interested in how different temperatures affect levels of aggression in participants. In this particular study, systematic variance would be caused by

• error variance
• participants’ current levels of stress
• different temperatures
• a mistake in the research design

Which of the following is true about test-retest reliability? 

• It measures how related the items on a test are
• It can only be done with measures that should remain the same over time
• It can only be done when you have two or more raters observing participant behavior
• It measures how valid the items on a test are

When testing test-retest reliability, researchers generally want two administrations of the same test to….

• correlate at least .05.
• correlate at least .50.
• correlate at least .70.
• correlate 1.00.

Which of the following scenarios best demonstrates predictive validity?

• A test for college admissions accurately predicts students' future academic performance.

• A test consistently yields the same scores when administered on two different occasions.

• A test accurately measures a specific trait, as confirmed by experts in the field.

• A test shows a high correlation with another test measuring a related construct.

Which of the following is true about face validity? 

• It is the weakest type of validity
• It examines if a test is measuring what is seems like it is measuring
• Just because a test has face validity does not mean that the test is actually valid
• All of the above

I decide to make my own anxiety questionnaire. I have participants fill out my anxiety questionnaire AND an already validated questionnaire, to make sure the correlation between the two measures is high. This demonstrates…

• Convergent validity
• Concurrent validity
• Discriminant validity
• Face validity

On a subjective well-being scale, people answer the same question worded in multiple ways. Researchers first evaluate whether people responded consistently to each item, despite the different wordings. This means that researchers are assessing

• Convergent validity
• Interitem reliability
• Interrater reliability
• Concurrent validity