To include or exclude "Not Applicable" responses?

Moderators: statman, Analyst Techy, andris, Fierce, GerineL, Smash

skinfreak
Posts: 13
Joined: Fri Apr 04, 2008 8:21 pm
Location: Pontypridd, Wales

To include or exclude "Not Applicable" responses?

Postby skinfreak » Sun Apr 06, 2008 10:39 pm

Hi all,

I have several questions in a 7 point scale which range from "Completely Disagree" to "Completely Agree". In addition, I have an option of "Not Applicable". Should I exclude this for reasons of calculating the mode? It would be fair to me to state that value 6 "Agree" is most prevalent for those responses that ignore the N/A option, but is this statistically right?

I don't want to deliberately skew the response, but at the same time for those respondents who the question does apply, the result would not come out as a fair mode.

And so, should I add the 8th value to the "missing" values in order to exclude this data from the results. Or is there another way of dealing with it?

Thanks!
statman
Administrator
Posts: 2737
Joined: Tue Jun 12, 2007 12:08 pm
Location: Florida, USA

Postby statman » Mon Apr 07, 2008 12:41 pm

If the question isn't applicable then why would you want it included in any analysis except for informational purpose tables, e.g. frequency distributions of the responses?
See the note below

NOTE: Please read the Posting Guidelines and always tell us your OS, the SPSS version and information about your study and data!

Statman
Statistical Services
skinfreak
Posts: 13
Joined: Fri Apr 04, 2008 8:21 pm
Location: Pontypridd, Wales

Postby skinfreak » Mon Apr 07, 2008 7:29 pm

I read conflicting things all over the place.

To put it simply, if I ask "what is your favourite colour: Red, Orange, Yellow, Green or Blue" but also put in "Don't know", some would argue that "Don't Know" is important. While I might find out that Yellow is the most popular colour fomr the respondants, this might be skewed if I include the Don't Knows.

I might be talking crap, but I am not sure the validiity of leaving out the don't knows.
statman
Administrator
Posts: 2737
Joined: Tue Jun 12, 2007 12:08 pm
Location: Florida, USA

Postby statman » Tue Apr 08, 2008 12:47 pm

This is a very different example, here, I agree, the DNK is a valid response. However, "Among those who had a favorite color, ..." would omit DNK.
See the note below

NOTE: Please read the Posting Guidelines and always tell us your OS, the SPSS version and information about your study and data!

Statman
Statistical Services

Who is online

Users browsing this forum: No registered users and 1 guest

cron