# Survey data analysis

4 Sep 2007 - 1:46pm

5 replies

1774 reads

I am currently using Excel to analysis a survey - mostly based on likert

scale. Can anyone recommend a tutorial/books etc that discuss survey data

analysis? I have found the mean, frequency, percentages etc..but would like

to know other measures that can be used to analyze the data.

Thanks,

Oliver

## Comments

Are you doing quantitative survey analysis for end-user research,

segmentation, and persona creation?

On 9/4/07, oliver green <oliverhci at gmail.com> wrote:

>

>I am currently using Excel to analysis a survey - mostly based on likert

>scale. Can anyone recommend a tutorial/books etc that discuss survey data

>analysis? I have found the mean, frequency, percentages etc..but would

>like

>to know other measures that can be used to analyze the data.

>

>Thanks,

>Oliver

>________________________________________________________________

>Welcome to the Interaction Design Association (IxDA)!

>To post to this list ....... discuss at ixda.org

>List Guidelines ............ http://beta.ixda.org/guidelines

>List Help .................. http://beta.ixda.org/help

>Unsubscribe ................ http://beta.ixda.org/unsubscribe

>Questions .................. list at ixda.org

>Home ....................... http://beta.ixda.org

>--

~ we

-------------------------------------

n: will evans

t: user experience architect

e: wkevans4 at gmail.com

-------------------------------------

Pick up an old marketing research text from a used book store or half price books ( http://www.halfpricebooks.com ). The basic analysis of this kind of data has not changed in a while so you should be able to get what you want for around ten bucks.

Mark

On Tuesday, September 04, 2007, at 02:46PM, "oliver green" <oliverhci at gmail.com> wrote:

>

I am currently using Excel to analysis a survey - mostly based on likert

>scale. Can anyone recommend a tutorial/books etc that discuss survey data

>analysis? I have found the mean, frequency, percentages etc..but would like

>to know other measures that can be used to analyze the data.

>

>Thanks,

>Oliver

>________________________________________________________________

>Welcome to the Interaction Design Association (IxDA)!

>To post to this list ....... discuss at ixda.org

>List Guidelines ............ http://beta.ixda.org/guidelines

>List Help .................. http://beta.ixda.org/help

>Unsubscribe ................ http://beta.ixda.org/unsubscribe

>Questions .................. list at ixda.org

>Home ....................... http://beta.ixda.org

>

>>>

Are you doing quantitative survey analysis for >>end-user research,

segmentation, and persona >>creation?To measure attitudes based on demographics - ideally I would like to

create 4-5 personas based on these in order to drive UI design

decisions.

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Posted from the new ixda.org

http://beta.ixda.org/discuss?post=20076

Oliver,

I'll start with very basics first..

Typically in such analysis one is trying to find major similarities

and contrast. If this data is over a period of time you would also be

looking at trends.

A typical Survey is structured as follows:

1. Screening criteria questions - these questions helps ensure the

correct audience answers key questions further on

2. Concept Intro - whatever you are trying to test is introduced

here, typically you'll mix it with something else also to remove bias

and get a comparative analysis

3. Measure - here is where most likert and other scale questions come

(ranking, constant sum, staple scale etc..)

4. Addition question - like more info on demographics

With these,you can start create cross tab reports - essentially you

put one variable (answer or combination of answers) in rows and other

on columns to how one impacts other. If you ahve structured your

survey correctly you would have some independent and some dependent

variables. http://en.wikipedia.org/wiki/

Dependent_and_independent_variables

mostly you plot dependent against independent.

Next thing is to see the key trends, similarities and differences you

see are significant or not - is it something you'll base your

decision or not. So you do a statistical significance testing. There

are simple ones like t-test, z-test, etc.. and more sophisticated

ones like Hypothesis testing. Just search for these terms -

statistical significance testing, t-test, z-test, Chi-square, ANOVA,

Hypothesis testing, Conjoint Analysis.. these should lead you to good

reading material on what kind of analysis to conduct.

Here is t-test - http://en.wikipedia.org/wiki/Student%27s_t-test

I am not familiar with a cheap enough software yet, which could help

you in all this, might be others in the group are.

On Sep 4, 2007, at 2:46 PM, oliver green wrote:

>

I am currently using Excel to analysis a survey - mostly based on

>likert

>scale. Can anyone recommend a tutorial/books etc that discuss

>survey data

>analysis? I have found the mean, frequency, percentages etc..but

>would like

>to know other measures that can be used to analyze the data.

>

>Thanks,

>Oliver

>________________________________________________________________

Some prefer Expert Choice. You can try it out from

http://www.expertchoice.com/. The software adapts a method called "Analytic

Hierarchy Process". Also it has a module for analyzing survey data. I'm not

sure if it fits to your needs but could be a good start.

Link: http://www.expertchoice.com/

Link: http://thequalityportal.com/q_ahp.htm

Bests.

Etkin Ciftci

PhD. Student, Dept. of Industrial Design

Istanbul Technical University

Contact Information

e-mail: etkin.ciftci at itu.edu.tr

On 9/4/07 10:00 PM, "Mark Schraad" <mschraad at mac.com> wrote:

>

Pick up an old marketing research text from a used book store or half price

>books ( http://www.halfpricebooks.com ). The basic analysis of this kind of

>data has not changed in a while so you should be able to get what you want for

>around ten bucks.

>

>Mark

>

>

>On Tuesday, September 04, 2007, at 02:46PM, "oliver green"

><oliverhci at gmail.com> wrote:

>>I am currently using Excel to analysis a survey - mostly based on likert

>>scale. Can anyone recommend a tutorial/books etc that discuss survey data

>>analysis? I have found the mean, frequency, percentages etc..but would like

>>to know other measures that can be used to analyze the data.

>>

>>Thanks,

>>Oliver

>>________________________________________________________________

>>Welcome to the Interaction Design Association (IxDA)!

>>To post to this list ....... discuss at ixda.org

>>List Guidelines ............ http://beta.ixda.org/guidelines

>>List Help .................. http://beta.ixda.org/help

>>Unsubscribe ................ http://beta.ixda.org/unsubscribe

>>Questions .................. list at ixda.org

>>Home ....................... http://beta.ixda.org

>>

>>

>________________________________________________________________

>Welcome to the Interaction Design Association (IxDA)!

>To post to this list ....... discuss at ixda.org

>List Guidelines ............ http://beta.ixda.org/guidelines

>List Help .................. http://beta.ixda.org/help

>Unsubscribe ................ http://beta.ixda.org/unsubscribe

>Questions .................. list at ixda.org

>Home ....................... http://beta.ixda.org