Data Analysis & Reporting Flashcards

A district administrator notices that the mean SAT score increased from 1050 to 1080 after implementing a new curriculum. To determine if this change is statistically significant, the administrator should:

Answer: Conduct a hypothesis test with appropriate controls

Statistical significance must be determined through formal hypothesis testing, not by observing the direction of change alone.

In data reporting, a misleading graph often occurs when the y-axis does not start at zero. This technique can make small differences appear:

Answer: More dramatic than they actually are

Truncating the y-axis magnifies visual differences, making small changes look much larger than they are in absolute terms.

A teacher wants to identify students whose scores fall more than two standard deviations below the class mean. In a normal distribution, approximately what percentage of students would this include?

Answer: 2.5%

In a normal distribution, about 2.5% of values fall more than two standard deviations below the mean (the lower tail only).

Which type of validity evidence is gathered by examining whether a test's content adequately represents the domain being assessed?

Answer: Content validity

Content validity is established by evaluating whether the test items comprehensively and representatively sample the intended knowledge or skill domain.

A researcher uses data from all students in a school rather than a sample. This approach is called:

Answer: A census

A census collects data from every member of the population rather than a subset, eliminating sampling error.

Two variables have a correlation of r = -0.92. This means that as one variable increases, the other tends to:

Answer: Decrease strongly

A correlation of -0.92 indicates a strong negative relationship, meaning as one variable rises, the other reliably decreases.

A school reports that the average class size is 28 students, but half of all classes have 20 or fewer students. This suggests the distribution of class sizes is:

Answer: Positively skewed

When the mean (28) is higher than the median (≤20), the distribution has a right (positive) skew caused by a few very large classes pulling the mean up.