1
$\begingroup$

I have a repeated measures design in which I tested multiple individuals with three different sounds (playbacks). The playbacks were 20 seconds in length and I recorded 1) how individuals responded (ie head turn left, head turn right, crouch, etc), 2) how long it took individuals to respond, and 3) whether their very first response to the sound (given they responded at all or with a directional response - turn of the head) was towards the speaker or not.

I coded a turn in the direction of the speaker as 1 and if they turned away from the speaker as 0. I also have a "silence" period prior to playbacks, for which I also recorded any movements of the head or body and the direction in which they turned.

I would like to be able to test whether individuals were significantly more likely to turn in the direction of the speaker, when they heard a sound, regardless of the sound they were played. Is there a way to do this or do I need to use the three treatment groups or use some sort of a "dummy" treatment group?

Improve this question
$\endgroup$
7
  • 2
    $\begingroup$ "I would like to be able to test whether individuals were significantly more likely to turn in the direction of the speaker" -- individuals more likely ... than what? What is the comparison you seek to make here? Are you trying to compare the propensity of the different sounds to produce turning? Something like "$H_0$: the probability people will turn in response to sound is the same for the three sounds" vs "$H_1$: the probability people will turn in response to sound is not the same for all three sounds"? $\endgroup$
    – Glen_b
    Jan 24, 2015 at 0:59
  • 2
    $\begingroup$ Were individuals exposed to all three sounds or only to one sound? If they were exposed to all three, was the order randomized? $\endgroup$
    – Glen_b
    Jan 24, 2015 at 1:02
  • $\begingroup$ Individuals were tested with all three sounds and playback of sounds was randomized. The comparison, I guess, would be whether they turned in the direction or in the opposite direction of the speaker or away from the speaker. Ultimately, I just want to be able to say something about their ability to recognize which direction the sound was coming from. $\endgroup$
    – igor76
    Jan 24, 2015 at 1:43
  • $\begingroup$ You can't find out that information if what you recorded was "yes/no" as you describe in your question. Please clarify your question on what information you have, and include the details just revealed. They're important in decided what analysis to make. $\endgroup$
    – Glen_b
    Jan 24, 2015 at 1:59
  • $\begingroup$ Ok, I guess that was my original concern. So I can really only report the proportion of individuals that responded in the direction of the speaker? The experiment was designed to address a different question, but this was additional measurement I took. I wasn't sure if there was some way to support that they can localize sound. $\endgroup$
    – igor76
    Jan 24, 2015 at 3:10

0

Reset to default

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.