SEM Series Part 4: Convert Continuous to Categorical

in this video I'm going to show you how to convert two continuous variables into categorical ones and why would even why would we even want to do that well we have two multi group moderators ones called frequency ones called experience and we want to see for those with low frequency what are the relationships and those with high frequency what are the relationships same with experience but right now if you look at the data set experience and frequency are not in a low and high category they're in a continuous set of values between zero and we got 13 here so what we need to do is find out what the medians for these two are and then create a low and a high that are on either side of that now those are relatively low relatively high rather than absolutely low and I okay so let's do that go to descriptives frequency go grab experience and frequency statistics what do we need we need the median hit continue and ok and we see that the median is three for experience and five frequency we can even look down here to see for experience we have roughly wow three and under is 54% and then four that's experience scuse me for frequency we have once per week and under is 51% 41.8% so we'll just use three and five for experience and frequency to split them will go to transform recode in two different variables go grab these guys here yeah experience in frequency for experience we're gonna rename it to experience Katz how about that and change and for frequency Rick cat cat for category change you can name it whatever you want you can aim at Bob if you want to all right old and new values for experience the value is 3 so lowest value through 3 gets a new value of 1 add and all other values get a value of 2 ad continue I mean be the same for frequency old new values lowest through like you know I'm going to click here and remove move and say lowest through 5 gets a new value of 1 add and continue and hit ok let's see if that did it properly I've never done two together actually it's run frequencies not on these two but on their cats okay and looks like it worked for I thought that might happen what it is was it took the most recent criteria and rules and applied it to both so it looked for 5s on expects cat expert cat head instead of threes let's go change that one back frequencies high frequencies apologize transform recode into different variable get rid of that and that one worked this one has got to be all the new values instead of yet see there it is five and move that make it out 3-1 ad so lesson learned got to do one at a time if there are different values okay let's go run the frequencies one more time on these two variables and here we have it 54% in 51.8% that's great now let's go save that data and that is how you create new variables you categorical variables based on continuous very data we can also do this real quick go to the variable view and in the label on a label in the values click on that and say a value of 1 is equal to low audio 2 is equal to high okay and then I can just copy control-c control-v that right ctrl-c ctrl-v it's not gonna let me do that seriously dropsy trophy no well let me do it so I'll do it here too all right 1 equals low 2 equals high square it works before and okay great and then save it and there you have it

Channel: James Gaskin

Views: 40,923

Rating: 5 out of 5

Keywords: Screencast-O-Matic.com, Statistics, SEM, SPSS, Variables

Id: fkRZ3whPioQ

Length: 4min 51sec (291 seconds)

Published: Thu May 02 2013