Klinny, you can use Chi-Square test of independence for your second model.
hi, i am doing a study on how color contrast and layout value effect the overall presentation of a screen.
and this is one of the null-hypotheses:
i. Null-hypothesis I: Layout value does not effect the overall presentation of the screen. (where the layout value was calculated based on few formulas, and the overall presentation of the screen was obtained from the 10 respondents' rating)
and for this, i'm using t-test for a correlation coefficient to evaluate the hypothesis.
for the second one,
ii. Null hypothesis II: There is no dependency between legibility and color contrast. (where the color contrast of the screen was calculated from formulas, and legibility was obtained from respondents)
i would like to know if i can test this hypothesis using chi-square test of independence?
for eg: the screens with medium color contrast, 5 respondents rated it as very easy in terms of legibility, 27 rated as easy, and so on..
Color contrast Very easy ..Easy .....Hard...Very hard
medium .........5 .............27 ........128....35
high.............64 1...........63 .......317 .....41
and based on the information above, we can decide whether there is enough evidence to reject the null-hypothesis.
i've gone through some readings on hypothesis evaluation, and it mentioned that it has to be both categorical variables to perform chi-square.
I'm just confused whether can i use chi-square in the second example? if not, which evaluation method should i employ?
thanks in advance.