Hot answers tagged

6

It's important to distinguish between measures and analyses, because only analyses can be quantitative or qualitative, not measures. Measures are, essentially, systematic processes by which we acquire our data, and analyses are processes we use to look at the data. As a rule of thumb, the difference is not hard to find and is given in the name: ...


6

ANOVA and t-tests are statistical tests for significance and therefore quantitative. The other mentioned items are scales (adding numbers to a certain choice) and therefore they can be considered as ordinal scales, and hence as quantitative as they are based on numbers. The NASA one can be administered by using a sliding scale which can be considered to ...


5

It depends on the types of changes you are looking to make. In general, my experience has been that changes to a response scale can be done with minimal threats to validity (e.g. looking to change the 4 item Likert type scale to a 6 item Likert type scale). On the other hand, changing question wording itself usually can't be done "without losing validity" ...


4

Here are a couple of suggestions: Inquisit records the timing of each response and can be run online. If you care about timing a lot and you need to run the study online, then this is quite a good option. Qualtrics can record various features of survey delivery time. In particular, you can record time spent on each survey page. If you have one question per ...


1

A few options. You could start with an open ended question and then move to closed ended questions for leading brands and perhaps any other options listed in the previously open ended process. You could use an input box that takes text and shows partial matches which people then select (e.g., a bit like in Facebook when selecting a friend)


1

it seems that you can use randomization of options to decrease bias. That is, present every respondent with multiple choices but in completely random order. Also, bias might be directly tested on two groups, say A&B. Only difference will be the order of available answers. This manual seems to be very helpful: Handbook of Survey Research, Ch.9 Question ...


1

The best way to set up a Mechanical Turk longitudinal study is with TurkPrime.com. You can follow-up with users and invite them to take a new survey. As long as they are properly incentivized they will come. I recommend that you inform them at the start that you are running a 10 part study and that those who perform well on each of the 4 minute studies will ...


1

Question: if you have used such softwares to make experiments, what did you think about the process? We use Paradigm (www.paradigmexperiments.com) in our lab and like it quite a bit. You can build most experiments just with the drag and drop builder interface. It also supports Python scripting just to give you more flexibility (and piece of mind in ...



Only top voted, non community-wiki answers of a minimum length are eligible