Nesstar Preparing the Dataset using SPSS

Hi there! This tutorial will show you how to prepare your dataset in SPSS. This requires you to have some familiarity with SPSS. Changes can be made in Nesstar, but these are limited, so it is recommended to first do it in your SPSS file, before we import into Nesstar. So using a file that I have worked on on previously, which is the Canadian Internet Use Survey, 2012, Person file. I will show you how to prepare your dataset. First, check to make sure there is an ID variable. We're going to do this by expanding the label and we can see that our first variable is the ID variable. Check, that's done. The next two steps are often the lengthiest part of working with your initial SPSS file. You will need to check that your Label and Values are correct and each one is unique to the other. Labels are limited to 40 characters and Values to 20. This can be counted quickly by copying and pasting into a Word document and using the Word Count function. So you would simply, right click, click copy, and throw it your Word documents or use keyboard shortcuts to make it even faster. Once that's done, bring it back into your SPSS file. Also, the use of abbreviations is encouraged. For example, respondent can be shortened to R. We see that here with Characteristic of community where R lives. It really means Characteristic of community where Respondent lives, but is shortened. Another example, household can be shortened to hhld. Sometimes it is difficult to shorten a label. If this is the case, an accompanying document can outline uncommon abbreviations. This can be added in Nesstar Publisher under Other Materials. You can also add cbk to the label itself to refer the user to the codebook or User Guide. Copy and paste again becomes your friend when looking at surveys in the same series as the same questions are often repeated. If you're stuck, just look at a previous year. Once labels are value labels are correct, you are going to need to declare your Missing values. If we go over to PROVINCE, highlight Values, we can see this little blue box. Click on that and it opens up your Value Labels menu. Scroll down to the bottom to see if there is any missing values. We can see it does with 96, 97, 98, and 99. Other common labels are Skip, Refused, or Don't Know. So remember those numbers, exit out of this menu, and highlight missing in the same way we did with these values. Click on the blue box. Now you have two options. You can either enter them individually. This can be done if you have a missing value of 9 and a missing value of 13. You would enter them here. Or for this one, we have a range of 96 to 99, so you would just have to enter 96 to 99 and click OK. By looking down we can see that all of our missing values have been labelled. Finally, ensure the data are is not weighted. This is done in Nesstar Publisher. Then you will save your file on your computer where you want it and you're done. This video was recorded by the Carleton University Data Centre. Thanks for watching!