Hi Kelly<div><br></div><div>This is a debate ongoing - technically it is all ordinal data (is 4 twice as much as 2?) but in most instances it is acceptable to treat iit as interval data,  especially when dealing with sums of data. You should be able to do parametric tests on this as far as I can see (although I am no expert)</div>

<div><br></div><div>-- <br><div>Many thanks</div><div><span style="font-family: Arial, sans-serif; ">Ben Haysom-Newport BSc (Hons), MBPsS, MSc</span></div><div><br></div><div><br></div></div><div><br></div><div><br></div>

<div><br><div class="gmail_quote">On 12 April 2011 11:51, Kelly Tate <span dir="ltr">&lt;<a href="mailto:[log in to unmask]"><a href="/cgi-bin/wa-jisc.exe?LOGON=A3%3Dind1104%26L%3DPSYCH-POSTGRADS%26E%3Dquoted-printable%26P%3D1571912%26B%3D--001636e0a4c8387aa004a0b6d751%26T%3Dtext%252Fhtml%3B%2520charset%3Dwindows-1252%26pending%3D" target="_parent" >[log in to unmask]</a></a>&gt;</span> wrote:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex;">

<div lang="EN-GB" link="blue" vlink="purple"><div><p class="MsoNormal">Hi,</p><p class="MsoNormal"> </p><p class="MsoNormal">I was hoping to pick the brains of anyone familiar in analysing questionnaire data. Basically I have created a questionnaire to determine attitudes towards packaging and have been rifling through various textbooks but cannot seem to get a straight answer! Responses are all on a bipolar 5-point scale with a midpoint (similar to a likert scale) thereby qualifying as interval data, with the exception that for some questions, each number has an assigned category, rather than being purely numerical, e.g.</p>

<p class="MsoNormal"> </p><p class="MsoNormal">1) I always purchase non-packaged items</p><p class="MsoNormal">2) I mostly purchase non-packaged items</p><p class="MsoNormal">3) I purchase both non-packaged and pre-packaged items equally</p>

<p class="MsoNormal">4) I mostly purchase pre-packaged items</p><p class="MsoNormal">5) I always purchase pre-packaged items</p><p class="MsoNormal"> </p><p class="MsoNormal">Does anyone know whether items using this format should this by analysed via parametric or non-parametric (i.e spearman’s rank) tests as I have had mixed responses?!  </p>

<p class="MsoNormal"> </p><p class="MsoNormal">Additionally, further questions are scaled using the terms ‘always, sometimes, frequently, rarely, never’- again these data fall between ordinal and interval as the distance between items is somewhat meaningful, should these be analysed using the same tests as the question above? </p>

<p class="MsoNormal"> </p><p class="MsoNormal">I am looking for a) the relationship between questionnaire items and b) the relationship between self-reported attitudes/behaviour and implicitly measured data ,</p><p class="MsoNormal">

 </p><p class="MsoNormal">Hope that makes some sense!?</p><p class="MsoNormal"> </p><p class="MsoNormal">Thanks in advance</p><p class="MsoNormal"> </p><font color="#888888"><p class="MsoNormal">Kelly</p></font></div></div>

</blockquote></div><br><br clear="all"><br><br>
</div>