Dividing a variable into categories in SPSS

hi there this is Muzammil and the question that we are going to answer today is divide the respondents into three groups depending on the level of satisfaction in life high medium and low take only the high and low groups for further analysis this is question number 4 and the gist of this question is to divide a particular variable which in this case is total satisfaction into three groups low medium and high and then we have to take only two variables low and high for further analysis how do we do this it's very simple first of all to understand what kind of data we have the range of the data let's run a simple analysis that's frequency steps click on analyze descriptive statistics and frequencies here we'll take into consideration this tote sat variable which is the total satisfaction we click here and then we click OK now let's take a look at the output in the output we can see that we we have a range which is between 23 which is the lowest level of satisfaction and 60 which is the highest level of satisfaction now this is what we need to divide into three groups low medium and high we can do that on the basis of the range we have here but that would not be perfect there will be so many mistakes with that a better way of dividing this data into three is to use cumulative percentage and divide on the basis of tests so we'll divide on the basis of tests but we need to find out that point which marks the dividing line between two groups how do we do that let's click on Windows and go back to our our data here we click on analyze descriptive in now we go to statistics here and click percentile now 33 percentile is approximately one-third of the completed asset we have in 66 and then hundred okay okay let's take a look at the output what it tells us okay so 33 percentile is at 40 so this is 33% time 66 is at 46 4666 and 60s at hundred so now we have the dividing lines between three different groups we know where to divide them once we have decided that in this case it's 40 46 and 60 which are the marking points let's go back to our data view here first of all we have the totes at variable we don't want to fiddle with the original one so we'll create one more we go to transform compute variable and here we write name for another variable totes at one we put it here and we click on ok now let's take a look okay here we have thoughts at one which is exactly a replica of thoughts at now let's fill it with thoughts at one and divide it into three divide offset one into three groups based on level of satisfaction of the respondents let's click on transform then re code in the same variable once we do that select or set one click on old and new values now here we know that a 33 percent Isle was at 40 so we'll write range we will write range lowest through 40 so which means from 0 to 40 would be called one that's the west now from 40 point 0 0 0 0 1 through 46 we'll call it 2 which is medium and then from forty six point zero zero zero 1 through highest we call it 3 that's high now we had done this click on ok let's take a look now we have divided it into three groups let's take a look at this variable now here 1 2 3 indicating high low and medium levels of satisfaction not prospectively so we had divided it into three now the second part of the question is first let's check in terms of data output how it works first we'll give labels to what set 1 we said one is low 2 is medium and three is high because okay run the same test once more okay oops see there's a mistake without to analyze descriptives and frequencies of we have to select dots at one that was a state women and click on okay let's take a look we had divided thoughts at one into three groups low medium and high low has 118 frequency medium has 114 and high has 117 this is the percentage now that we have done this is the second part of the question is to take only low and high levels of satisfaction for further analysis how do we do that we go back to the data View window here and we have to review this to remember what we do is that once again so as not to lose the data in talks at one will create one more replica weight variable how do we do that the same procedure here we call the new variable thought back to and okay okay here we have got set - which is a replica thoughts at one now we are going to work with dots that - and remove medium category which is - how we do that we select this and then we click on transform recode into same variable we remove this and add dots at - all the new values we remove all of these first now value to which we need to replace should be the new value for it should be system missing we click that we click OK and that's it we are done the job now let's run the test once more let's see what we have we select dots at - click click on ok and there we go one and three one means low three means high or we can give labels to this as well one and three like we did before one would be low three would be high we click on okay that's it now let's run the test again once more to see the changes okay lo hi so in this tutorial we learnt how to divide a set of data into three different levels high medium and low and how to take only high and low for further analysis this was a tutorial by Muzammil thank you for watching bye bye

Channel: StatArena

Views: 36,115

Rating: 4.8677688 out of 5

Keywords: data, groups, spss, divide, teaching, interval, nominal, category, categories

Id: X71m4_OvokU

Length: 8min 30sec (510 seconds)

Published: Wed Feb 12 2014