NotesFAQContact Us
Search Tips
ERIC Number: ED442853
Record Type: Non-Journal
Publication Date: 1999-Nov
Pages: 24
Abstractor: N/A
Reference Count: N/A
Contingency Table Statistics and Educational Reality: Problems with the Chi-Square Statistic.
Rozell, Diann S.
The Pearson chi-square test can be useful in situations in which the researcher wishes to compare observed versus expected frequencies in categories, or cells, or a contingency table. Although these tests can be useful, various problems associated with their use and interpretation are common. First, the chi-square test is often the result of weak research questions. Second, chi-square tests are often the result of weak or erroneous information about data. An educational research data set is used to illustrate that statistically significant chi-squares often do not inform the researcher about the contributions of the cells in the contingency table, resulting in unclear conclusions or the use of additional statistical tests, neither of which is a promising alternative. (Contains 3 tables and 24 references.) (Author/SLD)
Publication Type: Information Analyses; Speeches/Meeting Papers
Education Level: N/A
Audience: N/A
Language: English
Sponsor: N/A
Authoring Institution: N/A