0
$\begingroup$

Following a previous question I made I'd like to state once more that this is my first attempt at analyzing data with SPSS.

I have encoded and transformed my data properly and I also used some of the basic analysis types like Frequencies and Descriptive and also Cross Tabs (reference: What are good basic statistics to use for ordinal data?)

I'd like to take it a step further and I was wondering what analysis types should I use next?

I've heard that the next step would be linear regressions and then maybe t-test. Although I still have to study a bit to understand what each of those do exactly, I'd be interested to hear more oppinions on what would be best to look into beyond the basics.

I realize it's a bit subjective but I'd like some tools that are widely used and their output has the best possible practical interest.

I also believe that the answer to that would be useful for anyone in my shoes, otherwise I would think twice before asking...

Thanks again!

EDIT: In case I was misunderstood. I provided the reference question above because I am looking for similar answers in the form of specific suggestions. Like the question above I too have lots of questions that have Likert Style responses which I recoded as 1-4. I already used frequencies and other basic statistics and I'm looking for further useful suggestions. Just one or two that you consider useful.

$\endgroup$
4
  • 2
    $\begingroup$ Although your current 2 respondents have made valliant efforts, this question is not really answerable without more info. Can you edit / rework this Q so that it asks a specific, answerable question? If not, this Q will probably need to be closed. Please see our FAQ. $\endgroup$ Dec 3, 2012 at 15:16
  • $\begingroup$ @gung I think the question is as answerable as the question I provided as reference. Which is still open with 24upvotes. He asks: what basic summary statistics are considered useful? I wanted to go a step further from the answers given (frequency analysis etc) and ask for a couple more useful statistics beyond the basics. I don't know how my question differs so much from the one given.. If you think I phrased it wrongly feel free to edit it. But judging by everyone's answers I think they understood it perfectly. I just didn't get a specific suggestion yet, although they were useful answers. $\endgroup$
    – xray1986
    Dec 3, 2012 at 16:30
  • 1
    $\begingroup$ I think the problem with just saying "Do this next!" is that there are many things you could do with the data next but each might be used to answer a different question. $\endgroup$
    – DanTheMan
    Dec 3, 2012 at 16:50
  • $\begingroup$ @DanTheMan Ok fair enough. I will not debate it further. Based on this logic I will accept PeterFlom's answer for now until I find some appropriate Questions to ask. Thx all for commenting. $\endgroup$
    – xray1986
    Dec 3, 2012 at 16:57

3 Answers 3

5
$\begingroup$

I think your question is sort of inside out. The right path is not to get data and then ask "what can I do with it?" it is to ask "What do I want to do?" then get data then do the analysis. The "What do I want to do?" should be in non-statistical language.

Even if you have your data before knowing what you want to do (an all too frequent occurrence) you should not be asking "What statistical methods can I do with this data?" but rather "What interesting questions can this data help answer?" Then you use the answer to these questions to choose an analytic method.

$\endgroup$
1
  • $\begingroup$ Thank you for your answer. I realize that this would be the most appropriate way to go. Since I already have the data I had 2 options: 1) Think about questions that this data can help answer. 2) Think about answers that particular methods of analysis give you and decide whether they make sense with my current data. The reason I chose to go with the second option was my lack of experience in asking the right questions. Sometimes reverse engineering works wonders. $\endgroup$
    – xray1986
    Dec 3, 2012 at 12:26
2
$\begingroup$

Clearly, you would want to frame your research questions and choose an analysis, but sometimes an awareness of how data can be analyzed suggests, by way of example, questions you can pose. A good place to start learning about SPSS and Statistics is Andy Field's text Discovering Statistics Using SPSS (3rd ED). Sage. You can learn more about the text here: http://www.sagepub.com/books/Book233183

Best,

Stephen Brand

(Statisticsdoc)

$\endgroup$
1
$\begingroup$

I think that the following reference would be worth your immediate attention

http://www.jerrydallal.com/LHSP/LHSP.htm

while this would be useful once you have more experience

http://www-users.york.ac.uk/~mb55/pubs/pbstnote.htm

The SPSS manuals are worth reading, particularly the brief guide, the core system's user guide, statistics base and ProgDataMgmt book.

$\endgroup$
1
  • $\begingroup$ Thank you! I will check both links you provided as soon as I find some time and get back to you. $\endgroup$
    – xray1986
    Dec 3, 2012 at 12:26

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.