Hi, Well the survey is for a project looking into ways the Internet can enhance learning. The first part of the survey asks pertinent questions about their current study/learning environment, such as "I have trouble finding library books at the right time" and "I see lectures as a major source of learning", and they specify their opinion in the 5 point range strongly agree --> strongly disagree.
The second part uses the same scale, but asks for opinions on a number of innovations for using the Internet in learning, in terms of whether it would be an improvement. Therefore, I want to find correlations between groups of factors in the first section and opinions in the second section. Hope that makes sense - are there some obvious ways to tackle this analysis? thanks. [EMAIL PROTECTED] (Simon, Steve, PhD) wrote in message news:<E7AC96207335D411B1E7009027FC284902A9AED2@EXCHANGE2>... > It's difficult to answer a question that is asked so generally. You might > try explaining to this group why you collected the data in the first place. > For the most part, it is typically to: > > 1. characterize a specific group of interest, > 2. compare two or more specific groups, > 3. discover a pattern among several variables. > > If your answer is, "because I had to do it for an assignment" then you need > to take a step back and ask yourself why would someone else might be > interested in the data you have collected. You might also seek feedback from > your teacher (or your boss if this was a work assignment). > > There may also be multiple objectives. If so, just specify the two or three > that are most important or interesting. > > Don't be bashful and don't be vague. The more information you can provide, > the better answer we can provide. > > Do keep your objectives realistic, of course. Both because of the limited > time you have and the small sample size that you have collected. > > Steve Simon, [EMAIL PROTECTED], Standard Disclaimer. > The STATS web page has moved to > http://www.childrens-mercy.org/stats > > > -----Original Message----- > From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED]] > Sent: Saturday, February 16, 2002 4:59 PM > To: [EMAIL PROTECTED] > Subject: Likert Scale Analysis - HELP! > > Hi there, > > I have recently done a 5-point likert-style survey (with 34 questions) > and got about 45 responses. > > I am not great at statistics, and have not studied it since high > school! > > Can someone please give me some advice on how to analyse the data? > > I am not looking to do a full-on smart-ass analysis, I am realistic of > the time I can allocate to get this done (a few days max). I *would* > like to examine correlations between pairs (or even better - groups) > of questions. > > The trouble I find with trying to learn stats is that all the books > tell me what to use, but not WHY I should use it. > > I have been told that it would be good idea if I used SPSS, but it > looks a little daunting - easy for them to say. I am pretty good with > Excel... I am not a lazy person, but I would love to plug the figures > into a custom-made spreadsheet and then just set it running. > > Assuming that I can get the variables set up in SPSS, what are the > functions to use? > > > thanks. > > > > Instructions for joining and leaving this list, remarks about the > problem of INAPPROPRIATE MESSAGES, and archives are available at > http://jse.stat.ncsu.edu/ > > > -- ================================================================= Instructions for joining and leaving this list, remarks about the problem of INAPPROPRIATE MESSAGES, and archives are available at http://jse.stat.ncsu.edu/ =================================================================
