Millisecond Forums

> I pretty much just wanted to know if /optionvalues always necessarily overwrites /options in the data printout of
> dropdown.Q1.response (or radiobuttons or slider, etc.)

Yes.

Regarding how to phrase conditional statements: Inquisit 4 allows for standard if ... else if ... else statements, e.g.

/ ontrialend = [ if (boolean) {expression; expressions; ...}
                     else if (boolean) {expression; expression; ...}
                     else if (boolean) {expression; expression; ...}
                     else {expression; expression; ...}; ]

Thanks for the super quick reply. Glad to know I wasn't missing anything obvious, except that, about two minutes after I posted the question, I realized I put "=" instead of "==" in my if statements. I think I'll use this as the basic format of the conditional statements:

/ ontrialend = [values.Q1rev = if(dropdown.Q1.response == 1) 4 else if(dropdown.Q1.response == 2) 3 else if(dropdown.Q1.response == 3) 2 else if(dropdown.Q1.response == 4) 1]

I actually have a broader question about using Inquisit to collect questionnaire data. I know its focus is on reaction time tasks, but it's also a really convenient substitute for pen and paper surveys. The lack of a scrollbar for longer questionnaires is somewhat inconvenient, but it still beats out the time investment of scoring results by hand, entering the data to electronic resource, and implementing multiple redundant steps to make sure everything is correct. On that note, are there any recommended "Best Practices" for making long surveys in Inquisit. I mean something like the BIS, which is 30 questions, or even the BRIEF-A, which is 75. Are there recommendations on the number of questions per page, how small the font can be, how often directions should be repeated, whether to use dropdown or radiobuttons, etc.? I've found using dropdown lets me make the font a little bit bigger. I'm not having difficulty trying things out on my own, but I would appreciate any general advice that comes from experience.

Thanks!

In the wild, available screen estate varies, i.e., you'll encounter systems running at vastly different screen resolutions, different (physical) display sizes, etc.

Useful rules of thumb, IMO, are:

(1) Express sizes and positions in percentages, including font sizes. For survey question items, you have pretty comprehensive control over font sizes etc., via the various fontstyle attributes.

(2) You can use the <defaults> element's /canvassize, /canvasaspectratio and /canvasposition attributes to achieve relatively consistent results across varying screen ratios.

(3) Questions like "How many items per page?", "How often to repeat instructions?", etc. are hard to answer sensibly in general terms. The answer ultimately depends on the type and complexity of the questionnaire you're implementing. When in doubt, I'd recommend pre-testing different variants using small samples.

Thanks for the advice! There is one thing I meant to ask about. Sometimes when I've copy/pasted the questions from word documents into Inquisit, I've been able to get way with having quotes within the text without using ~". It makes the inside quotes a different style than the normal " as well. I've attached a png of the phenomenon. Could this ever possibly create issues with the script? I haven't had an issue related to it yet.

This can happen if you copy / paste stuff from other applications and you should be careful with that. It will not cause problems in this particular instance, but it can in others. Quotes play an important role in Inquisit syntax, among other things, the parser uses them to determine where a given item starts and ends. If those quotes are of the "weird" kind, the parser won't be able to do its job and errors can result (such as not recognizing an item).

That is why one ought to escape quotes when using them within an item:

<text mytext>
/ items = ("She said ~"Hello!~" with a big smile on her face.")
...
</text>

For details, see the "Special Characters" topic in the documentation.