Social Philosophy of Science for the Social Sciences by Unknown

Author:Unknown
Language: eng
Format: epub
ISBN: 9783030330996
Publisher: Springer International Publishing

Checking the Reporting of Statistical Tests

While it is important to evaluate the overall claims of published research, simply checking the reported data can give an idea of the validity of presented research. Nuijten et al. (2016) presented statcheck, a tool to test whether the reported p-value matches the presented test statistics and degrees of freedom. The application has been found to show good reliability compared with a manual process of checking test statistics. Statcheck helps researchers keep track of rounding errors or misreporting. Unfortunately, the application is only able to recognize the most common tests at the moment. The statcheck package (Epskamp & Nuijten, 2014) can be used in the R environment or online (http://​statcheck.​io). Another way of discovering errors in reporting has been the GRIM test (Brown & Heathers, 2017). Basically, the test verifies whether a reported mean is plausible based on the reported sample size. Imagine two participants filling out a 5-point Likert-type scale. In such an example, a mean of 2.5 or 3 would be nothing out of the ordinary, but a mean score of 3.33 would be impossible based on the sample size. The GRIM test can point out such anomalies and is available online (http://​www.​prepubmed.​org/​grim_​test/​).

Download

Copyright Disclaimer:
This site does not store any files on its server. We only index and link to content provided by other sites. Please contact the content providers to delete copyright contents if any and email us, we'll remove relevant links or contents immediately.