A hypothesis involving one independent variable and one dependent variable is:
A simple hypothesis has one IV and one DV.
A test contains items scored only as correct or incorrect. Which reliability measure is particularly appropriate?
KR-20 is appropriate for right/wrong items.
Which reliability method is specifically designed for tests containing binary right/wrong items?
KR-20 is for binary-scored items.
Several researchers independently rate the same classroom behavior. Their agreement is used to determine:
Interrater reliability concerns agreement among raters.
What is the key distinction between parallel forms and split-half reliability?
Parallel forms use two versions; split-half uses one test divided.
A researcher divides a test into two sets and correlates performance on both sets. This procedure represents:
Split-half reliability divides one test into halves.
Two equivalent versions of a test are administered to the same respondents. This assesses:
Parallel forms use two equivalent test versions.
Test-retest reliability is especially suitable when the characteristic measured is expected to:
Stable traits are suitable for test-retest reliability.
The same test is administered to the same group at two different times. Which reliability is assessed?
Test-retest reliability repeats the same test over time.
A test produces highly consistent scores but does not actually measure the intended construct. The test is best described as:
Consistency does not guarantee validity.
An established external measure is used as a benchmark against which a new instrument is judged. This concerns:
Criterion validity uses an external benchmark.
A researcher asks whether a depression questionnaire actually measures the construct of depression. This primarily concerns:
Construct validity concerns measuring the theoretical construct.
A questionnaire appears appropriate for measuring a construct merely by examining its items. This represents:
Face validity is superficial appearance of measuring the construct.
A biology test excludes several major topics that students were taught. Which validity is most directly threatened?
Content validity concerns coverage of the domain.
A newly developed test is compared with an already validated test measuring the same characteristic. Strong correlation supports:
Concurrent validity compares a new test with an established one.
A study is carefully conducted, but its findings cannot reasonably be generalized to real-world settings. Which validity is most directly limited?
External validity concerns generalization beyond the study.
A researcher records students’ heights to the nearest decimal point. Height is most appropriately classified as:
Height can take continuous numerical values.
Students receiving a new teaching method also happen to receive substantially more parental support. Parental support may therefore act as a:
Parental support may confound the teaching method effect.