lecture 5 - analysis of qualitative data Flashcards
for evaluating qualitative frequency data (nominal or ordinal data) for ‘goodness of fit’ (with a hypothesized underlying distribution) or for ‘association’ between multiple measures
analysis of qualitative data
goodness of fit: we determine the significance of the difference between observed and expected results using _______ statistic
this statistic has positive values only
chi-square
for a chi-square with a degree of freedom LESS THAN 1, what is recommended
Yate’s correction (- 1/2 in chi-square formula)
true/false: if the df is >1, Yates correction is not required
true
true/false: random sampling is not essential for calculation of chi-square
true - but may impact the interpretation (bias) therefore ideally they would be random
true/false: observations must be independent for chi-square
true
true/false: categories must be mutually exclusive for chi-square
true
this is an assumption for chi-square; in all the cells their must be at least 1, with no more than 20% of cells with an expected frequency < 5 for the resulting theoretical distribution to be reasonably accurate
expected frequencies
the distribution of chi-square depends on the number of treatments being compared and the number of outcomes. this dependancy is quantified in a degrees of freedom parameter v equal to:
number of rows in the table -1 times the number of columns in the table - 1
v=(r-1)(c-1)