3
$\begingroup$

We have been designing a series of experiments on comparing different active reading techniques for college students. In most of the studies we found in the literature, the passages are arbitrarily chosen based on the courses that the researchers teach or the topics related to their subject pools.

Are there standardized reading comprehension tests that are reliable and generalizable for college level students? Unfortunately, we could not find any ACT, SAT, AP, or IB reading comprehension tests that are publicly released for research purposes. Passages that cover a variety of topics like humanities, social studies, natural sciences, and literary narratives and similar in length to the ACT long passages are what we are looking for.

If such standardized tests do not exist for researchers, how do you advise we generate such reliable and generalizable reading comprehension tests?

Improve this question
$\endgroup$

1 Answer 1

Reset to default
2
$\begingroup$

The Nelson-Denny reading test has a subtest containing 7 reading passages with normative data for individuals up to 24 years 11 months including 4 year college students with 36 comprehension questions.

Per the Pro Ed website: "Statistical Characteristics

Age-based normative data (N = 4,010) are provided for students ages 14 years 0 months (14-0) through 24 years 11 months (24-11) enrolled in high school, 2-year, or 4-year university programs. Grade-based normative data (N = 3,487) are provided for students of any age who are enrolled in high school, 2-year, or 4-year university programs. The normative samples were stratified to match 2017 U.S. population estimates. For both forms, average internal consistency (content sampling) reliability coefficients range from .85 to .95 (very large to nearly perfect). Average alternate-form reliability ranges from .88 to .94 (very large to nearly perfect). Average test–retest reliability ranges from .89 to .98 (very large to nearly perfect). Correlations with well-known reading measures range from .47 to .84 (moderate to very large). Diagnostic accuracy studies indicate that the NDRT is able to accurately identify students with reading difficulties (sensitivity = .73, specificity = .80, ROC/AUC = .88) with few false positives."

Improve this answer
$\endgroup$
2
  • $\begingroup$ What evidence is there indicating these tests are reliable and generalizable to topics covered at college level? $\endgroup$
    – user2521204
    Commented Jul 24, 2022 at 18:08
  • $\begingroup$ I'm new here so unsure how much detail to go into, but there are many papers since the test's development that use it on different populations, some favorable and some unfavorable usually regarding validity rather than reliability. Without purchasing the test, I can't tell you what specific topics are covered beyond what's on the Pro Ed website. $\endgroup$
    – nouveaulingua
    Commented Jul 25, 2022 at 14:29

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.