Recoding into Different Variables in SPSS

hello this is dr. grande welcome to my video on recoding variables in SPSS I have here a data view and I have 4 variables I have an ID number an independent variable two levels control and CBT cognitive behavioral therapy and then I have a pretest score and a post-test score and these are recorded in scale at the scale level of measurement so say I want to recode pretest to be ordinal low medium and high and I have a range of values for each of those ordinal levels so go to transform I'm going to recode into different variables I'm going to move pretest over into this list box and I'm going to name the new variable PT ordinal and then the label I'll make the same PT underscore ordinal so you're going to select change or click change here so you can see pretest into PT ordinal and then I'm going to here says old and new values and from first value I want the lowest possible value so it's range lowest through value as I'm selecting and I want to go up to let's say 37 I don't want that value to be low so then I'm going to move over here I'm going to change the output variables to strings make that notation and then in the new value I'm just going to type it low so all the values from the lowest value in the pretest variable up to 37 will now be recategorize Dazz low so I'm going to add that and now I'm going to create a range let's say well Oh 38 - let's say 44 again output variables are strings and I call this one medium remember to click Add so now we have lowest through 37 will convert to low 34 or 38 rather through 44 will be medium and then we can do range value through highest so it'll be 45 - the highest value and of course here I'll put high add that so I have low medium and high click continue and now you can see that you can click the ok button here on this recode in a different variables dialog click OK and you can see now it's associated with the pretest the medium the high and low based on this pretest value you can see your an example of 32 it's been converted low of 42 paper in a medium and here's a 45 this will convert to high so we can also look for a specific value and code that I want to show you how we can add that so I'm going to open the same dialog go into old and new values and I'm going to remove 43 45 through highest as high and remove that one and create a range 45 through 49 and set that one as the high value and then add a specific value of 50 let's just say this particular value demands a higher level of caution so I'll just call this one caution because this will be the highest score possible on on the pretest so I'm going to add that so 50 will change into caution the low value through 37 low 38 to 44 medium and 40 five through 49 now we'll convert to high click continue and ok and you can see now there was a value of 50 down toward the bottom of the data set and in the new variable it's now coded as caution so that those are the options you have from the transform and recode into different variables other than the missing variables you have the range lowest to the value value through the highest and then specify that you can also select all other values as well but this feature in SPSS is an efficient way to create a new variable and move it in this case from scale to ordinal it's important to note that you don't have to convert a value or a range of values into an ordinal variable as I did here you could also convert it into another scale level measurement for example you could take the value of 50 and you could recode that and you would you would unclick the output variables or strings and you code that to say 55 so you can you can move from scale to scale and similarly you could do the same thing with range now you could take a certain range say 44 through 49 and change that to some other numerical value say 47 so this will take this range and convert to 47 now it's important to be aware here that if you're starting with a scale variable where difference between each value was equal interval and you make a recode like this that you are going to be losing that equal interval property but you can complete that function in SPSS you can recode in two different variables scale variables ordinal or nominal if you are converting into a string though make sure you check off the output variables or strings box down here I hope you found this video on recoding variables in SPSS to be used as always if you have any questions or concerns feel free to contact me and I'll be happy to assist you

Channel: Dr. Todd Grande

Views: 180,828

Rating: 4.784348 out of 5

Keywords: SPSS, recode, recode into different variables, transform, Statistics (Field Of Study), nominal, ordinal, scale, equal interval, range, Grande

Id: K-eY-AXlERA

Length: 7min 31sec (451 seconds)

Published: Wed May 06 2015