0
$\begingroup$

In my study I have 2 Independent variable, first is situational factor and second is personality type (with 2 variations of type A and type B). My dependent variable is choosing the right answer, either A (correct, and scored as 1) or B (wrong, and scored as 0).

I have two questions. First, what type of data is my DV?. It might seems like categorical, but the data is more than just like gender of male or female. It has values of intelligent and accuracy in it. Second question is, which statistical measurement should I use?. Thank you so much for your kind answer.

$\endgroup$
  • $\begingroup$ Where, exactly, do "intelligent" and "accuracy" appear in the two possible values of "A" and "B"? $\endgroup$ – whuber Oct 26 '17 at 23:02
  • $\begingroup$ I provided the experiment participants with lot of information with cues for the right answer. If he/she manage to find those information then he/she will find the right answer. The ability to find and process the cues is what i mean by intelligence and accuracy. I have read examples for categorical, and they are more like something given, as i wrote above such as gender. I think my data is more than that. I am in uncertainty of claiming my DV as nominal, interval, ordinal or ratio. Thank you. $\endgroup$ – Hakeem Oct 27 '17 at 1:21
  • 1
    $\begingroup$ What is uncertain here? Your DV is a categorical (nominal) and binary variable. So, you have other variables such as intelligence and accuracy which are ordinal, interval, etc.? This has nothing to do with the scale of measurement of your DV. $\endgroup$ – T.E.G. Oct 27 '17 at 2:31
0
$\begingroup$

I'll presage this by saying that you're going to create problems for yourself by thinking too hard about this.

First, what type of data is my DV?. It might seems like categorical, but the data is more than just like gender of male or female. It has values of intelligent and accuracy in it.

Your intuition is correct, although your reasoning is not entirely sound.

The DV is dichotomous (binary), but it can also be viewed as an ordinal variable, because here, the value of 1 refers to the respondent possessing more of something. In this case, 1 means they have more knowledge/ability/proficiency.

data is more than just like gender of male or female.

Says who? Gender is traditionally viewed as a categorical variable, but if we define gender as the number of penises present, it's actually an interval scale.

Second question is, which statistical measurement should I use?.

You would need to elaborate more on what your study is. For example, for reasons noted above, personality type could be viewed as an ordinal scale where an individual increases in competitiveness or propensity to give themselves a heart attack (type Bii, Bi, Aii, Ai -- or whatever archaic personality theory you're going for)

Most likely, personality is a categorical variable and the study has a 2x4 design. In which case you'd perform a chi-square test.

$\endgroup$
  • $\begingroup$ You are right, I think too hard for this. Thank you Faustus. $\endgroup$ – Hakeem Nov 1 '17 at 1:45

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.