Quant: Lecture 8 Flashcards
Does nominal data contain quantitative data?
What about numerical?
No it contains categories.
Yes.
If you have have a nominal IV and a continuous DV, then what statistical test do you use?
Both variables continuous?
Both nominal?
t test
Pearson’s r
Chi
What’s the purpose of the chi squared test?
To test for significant differences in the pattern of responses between conditions.
How do you calculate chi squared?
Firstly, you put your results in a table. One row for each condition and 2 columns, each displaying a different type of response. The results are frequencies. Now, you find the total for each row, each column and an overall total of the frequencies collected. Fo is the observed frequencies, then you find the expected frequencies (Fe). To do this you do row total times the column total divided by 100 for each cell in the original table. Then you do Fo minus Fe for each piece of data, then square these results and then divide each of them by Fe. Then total all of the scores in the (Fo-Fe)2/Fe to get your chi squared value.
How do you work out the degrees of freedom for chi squared?
Number of rows - 1 times the number of the columns - 1.
Can you make directional predictions with nominal data?
Should the hypotheses be two-tailed?
No because the data is unordered
Yes.
How should you word your answer for chi squared?
The calculated X2 value (*) was er than the critical value () at a significance level of pt a significant difference between the IV and the DV. (You can also make conclusions about the observed frequencies on their own, just saying).
