 SIGNIFICANCE TEST The chi-square significance test in the far-right column measures the likelihood that the observed association between the independent variable (e.g.,'age') and the dependent variable (e.g., 'participation in the given activity') is caused by chance. Among statisticians a chi square of .05 is a conventionally accepted threshold of statistical significance; values of less than .05 are commonly referred to as "statistically significant." In practical terms, a chi square of less than .05 means that if, in fact, there was no association in the population between the independent and dependent variables, the observed association would be expected to occur by chance less than 5 times in 100 samples of the type we used. Thus, when the chi-square is less than .05, we can be confident in rejecting the possibility that no association exists between the independent and dependent variables. As the chi-square increases above .05 the likelihood that the observed association occurred by chance increases. NS indicates that the chi-square is not significant using the .05 threshold.
Home/List of Tables/Methdological Appendix