00:00:06:19 - 00:00:40:18 Narrator Welcome to the Variable Search tutorial in the NLS Investigator video series. This video will build upon information in the Getting Started video, and go more in depth about creating and customizing your search criterion, displaying search results, and provide additional tips to access the NLS Investigator for this tutorial, navigate to NLSinfo.org and click the Access Data / Investigator tab or visit the link on screen. 00:00:40:20 - 00:01:03:21 Narrator We will begin the Variable Search overview as a guest by clicking on the search link to enter the NLS Investigator tool. Once inside Investigator, I have chosen NLSY97 for the sample demo. If you do not see NLSY97, simply click on the drop-down menu below, select the study you want to work with, and choose from the selections. 00:01:03:28 - 00:01:27:04 Narrator Once the cohort has loaded, you will see five main tabs. For this demo, we'll be focused on the Variable Search tab. For more information about the other four tabs, view the Getting Started video in the NLS Investigator series. Once you click on the Variable Search tab, there will be three sub-tabs. The first is Browse Index, which contains the most commonly used variables. 00:01:27:06 - 00:01:59:19 Narrator This function does not access all of the cohorts variables, just the popular categories. You can access the different variables and topics by clicking on the plus and minus to expand and contract the list. Once you've found your topic, you can click on the variable to load it into the window on the right. The second sub-tab is Browse Index with Search, which allows the researcher to both browse the popular categories and use the Search drop-down menu to further refine the search criteria. 00:01:59:21 - 00:02:28:24 Narrator The third sub-tab is the Search without the Browse Index portion, and contains several options for searching for the variables. The Search sub-tab will give researchers access to all of the cohort variables during the search process. Note that not all cohorts will have the Browse Index or Browse Index with Search sub-tabs available. For this demo, we will primarily focus on the search sub-tab while in the search sub-tab area. 00:02:28:25 - 00:03:03:21 Narrator The first menu lists 8 main search categories that are available, such as area of interest, question text, question name, survey year, codebook, and many others. Some items allow the user to enter search terms, while others provide a list of topics to choose from. For example, if you choose area of interest, you will see that the second drop-down menu for logic operators and the third drop-down to refine the search criteria have loaded in. For your logic operators, 00:03:03:22 - 00:03:32:10 Narrator an area of interest you're given equals or doesn't equal. If you should select survey year, you will see 'equals', 'doesn't equals', along with 'within range'. For other options you'll have different logic operators as well. For instance, question name (enter search term), you have many more logical operators. Returning to area of interest, we'll leave the second drop-down with equals for the logic operator, 00:03:32:13 - 00:03:59:24 Narrator and then, because area of interest is a pick from list, there is already a predefined list for the user to choose from. For this, we'll go into education status and attainment. When the area of interest is loaded, you'll see a number of variables for that particular topic listed in the drop-down menu. You will also see a number of variables listed next to the Display Variables button. 00:03:59:25 - 00:04:28:28 Narrator This number will update depending on how many criteria you add to your search. To add multiple criteria, use the Add button. When you click the add button for your first criteria, it will pop up the search options for adding more criteria. For instance, you might want to limit your data to a particular survey year. For this example, we'll choose 2006, and you'll see that the number next to the Display Variables button has updated to 315. 00:04:29:06 - 00:04:59:24 Narrator This is because the operator above the criteria is set to intersecting, but you can also select all and you'll see that that number jumps to over 18,000 because there's 13,000 plus variables in 2006 and 5000 plus variables in the area of interest we have chosen. For this demo, I'm going to stick with intersecting variables and to execute the search, click the Display Variables button. To display results more efficiently, 00:04:59:24 - 00:05:31:26 Narrator Investigator groups variables and questionnaire loops referred to as variable display compression. For instance, the created variable CV Associates Credit is condensed into one line instead of four lines for each school. You can see the different school lines by clicking the plus sign to expand the rows, or clicking the minus sign to compress them. If you'd like to choose all of the variables in this loop, you can click the checkbox or tag the items. 00:05:31:26 - 00:06:00:04 Narrator And if you'd like to select only particular schools from this variable loop, you can expand the rows and only tag the items that you need, such as the first school line. For this example, I'm going to keep all the variables in this particular loop tagged for the demo. Note when you see your list of variables, you can access the codebook information by rolling over the RNUM or the question name. 00:06:00:09 - 00:06:23:00 Narrator This rollover pop-up menu is the codebook preview to see the full codebook information for each variable, either click the question name or click on the RNUM and it will take you to the Codebook tab to see all of the information. To learn more about the Codebook tab, view the Getting Started video in the NLS Investigator series. 00:06:23:02 - 00:06:53:16 Narrator To return to your variable search, click the tab and it will take you back to your current search settings. For this demo, I'm going to add a few more variables, such as highest grade and highest degree completed. Note I have since tagged six new variables that are listed in my Review Selected variables tab. When I first selected the NLSY97 cohort by default it listed required in recommended variables. 00:06:53:16 - 00:07:25:22 Narrator The required variable is the public ID or the identification code for each of the respondents. It also listed the gender, the birthday by month, birthday by year, the sample type, race and ethnicity. If I don't need any of these variables for my research, I can simply untag them. In this case, I'm going to keep the gender variable for demo purposes, and then if you click refresh, it will update your list of tagged variables. 00:07:25:25 - 00:07:53:25 Narrator Returning to the Variable Search tab, you will notice that there is also a tag all selection at the top bar. This will only tag all the variables on the current page. Right now I have 315 variables and they're all listed because my page limitation is set to 500. Our page limitations can be set to 1500 variables per page and 5000 variables per page. 00:07:54:00 - 00:08:25:24 Narrator This comes in handy if you have a large dataset similar to adding all possible variables, you can select 5000 for your page, and if for some reason you were to tag all these variables, you'd get the first 5000 by tagging all with the checkbox at the top next to RNUM. Another option for the number of variables that you see in the list is your primary variable filter, where you can also choose 'primary and secondary' variables or 'all variables'. 00:08:25:25 - 00:08:55:04 Narrator By default, the search will choose the primary variables which are the most important, generally include all created in event history variables, all roster variables, and respondents responses to survey questions that are not superseded by better information in a roster or created variable. For the primary and secondary variables, the secondary variables will add questions asked in the interview that are superseded by created a roster variables. 00:08:55:04 - 00:09:19:29 Narrator These questions may be useful for very detailed research about a narrow topic. They won't necessarily be needed by most researchers. The all variables includes tertiary variables, which are generally machine check items. These are included in the data to enable researchers to follow the flow of the interview, but in most cases they should not be used as input variables in research. 00:09:20:00 - 00:09:44:24 Narrator These are only available when the all variables variable preference level is selected. Using the Search tab and checking all variables will give you access to the maximum amount of variables for your particular search criteria, but in most cases, selecting the primary variables will be all that you need. Above the variable list on the left is the tab called Options. 00:09:44:26 - 00:10:20:20 Narrator Click on this tab to open or close the Options menu in the Options menu. You can select the variable characteristics you want displayed in the variable list by checking or unchecking each option. By default, the following options are already checked: the reference number, the question name, variable title, survey year, and compress all group rows. Other selections include the 'show area of interest column', which in this case, because we already have an area of interest selected in the search criteria, educational stats and attainment, 00:10:20:20 - 00:10:45:09 Narrator those are the data points we'll see first. We remove that topic and execute our search so that we just have all the variables in general from the survey year, you'll see the area of interest changes per topic for those variables. 00:10:45:12 - 00:11:02:07 Narrator The next item is to 'show variable type' column. This is where it will list machine checks, created variables, methodology variables, respondent responses, and so on. 00:11:02:09 - 00:11:46:12 Narrator The next is 'show valid number of responses'. This will show for the particular survey year how many valid responses there was for each variable. So in this case, age... month... as of interview date, there was 7,559 valid responses in 2006. For the hours worked per week at job 01, there were 6,704 valid responses. To get more details about the responses, you can rollover the question name to see the codebook preview, or you can click on them to see the full codebook page. 00:11:46:14 - 00:12:17:02 Narrator The next option is 'show variable preference and column' and this shows whether it's primary secondary, tertiary. Right now we have the filter set to primary so those are the variables that are showing up in that column. And then 'show data column' as a data column on the right of the variable list. When this option is checked you can enter a respondent's ID number to see their information related to each variable. 00:12:17:04 - 00:12:54:22 Narrator So if we start with ID 1, respondent 1 and click enter. When we scroll down, we will see the information for respondent 1 as it relates to the different variables. In this case 302 months at the date of the interview. Age at interviews 25. Census region was 1. So if you scroll over to see the codebook preview, you can see 1 is the northeast. 00:12:54:25 - 00:13:25:15 Narrator Current citizenship listed as 2, which is citizen but naturalized and so on. You can go down the column to see how respondent 1 data is reflected in those variables. One more option that will be useful is the 'show data column navigation controls'. So if you would like to navigate through the different respondents without having to type in their numbers, say you're going through a group of respondents consecutively. 00:13:25:22 - 00:13:50:18 Narrator the 'show data column navigation controls' will add the buttons to the right of the ID number box, so you can click the arrow and it will take you to respondent 2. And we'll see in 2006 that respondent 2 was a non-interview, a -5. You can click the arrow again to go backwards and forwards through the respondent IDs 00:13:50:20 - 00:14:24:02 Narrator using these navigation controls. Respondent 3 also a non-interview for 2006 and so on. One more note about the variable list, by default the variables are organized by RNUM, but you can organize them by any of the other columns that you have chosen through the Options menu, simply by clicking on the links, and it will reorganize those variables based on the column that you have selected. In this case, question name... 00:14:24:04 - 00:14:52:10 Narrator so then it starts at the biological adopted child roster. Now that we have covered the different display options for the variable list and tagged a few variables along the way, which can be seen in the Review Selected variables town, it's time to save or download the data. The first is to save or download the tagset for future reference, especially if you plan to return and add or delete variables from the criteria. 00:14:52:10 - 00:15:17:06 Narrator If you are an Investigator as a guest, you have the option to save to your PC. If you are registered user, you can save on the server. You have the option to save 'by RNUM' or 'by question name with year'. You enter a filename and click save. It will download the tagset in a text file with either the R number queue name with year designations. 00:15:17:08 - 00:15:45:14 Narrator The next tab is the Basic Download, which will provide a copy of the tagset, along with files for different statistical software such as STATA and SPSS and codebook information. For those of you using R, you will need to use the Advanced Download tab. In this instance, once again, you will use Filename and click download. In the Advanced tab, you can choose only the files you need and have the option for R. 00:15:45:16 - 00:16:14:08 Narrator In this case, if I am an R user, I will check on the source code and I might not need this STATA, SPSS and SAS files so I can uncheck those. One required file will be the tagset and other recommended files are the codebook definitions for the variables and the Excel file. Other options in the Advanced Download is a 'create frequency table', where you can make a simple table from your tagged variables, such as highest grade completed. 00:16:14:08 - 00:16:43:15 Narrator And because this is 2006 survey year data, I'm going to wait it with the 2006 weights. Click download. This will take you to the Managed Downloads tab, where you can see an HTML page with your frequency chart. Returning to the advanced download within that frequency table, you can also do a cross-tab, for instance the highest grade completed broken down by gender. 00:16:43:18 - 00:17:18:01 Narrator Once again choosing the 2006 weight option, clicking download, going to the Manage Download tab, and then viewing the corresponding HTML page. And you'll see it's now broken down by male (1) and female (2) of highest grade completed. The final item in the Advanced Download is the 'apply universe restrictor', and you can pop-up in the 'show equation editor' to see a list of your variables and the different operators to create your expression. 00:17:18:05 - 00:17:50:20 Narrator In this case, maybe my research is centered around the female respondents and I don't need the data from the male respondents. Instead of downloading all the data, I can limit it to just the females by clicking on the gender variable... equal... and I know females are designated as 2... add my expression. And then when I click the download, to create the data files that I need for my research, it will take me once again to the Manage Downloads. 00:17:50:20 - 00:18:16:27 Narrator This time it will now be an HTML file, it'll actually be a zip file. I'll click download. When I open that file, I didn't give my file name specific name, so they still say default, but I did choose R source code. It gave me my tagset for the NLSY97. This is a simple text file I can click and open. And I chose to download by RNUM. 00:18:17:00 - 00:18:29:13 Narrator And then when I open the Excel sheet, you'll see that my gender variable is restricted to only females from my universe selection. 00:18:29:15 - 00:18:40:09 Narrator Additional variable search tips, including common search problems, are available in the tutorial section of the NLS website. 00:18:40:12 - 00:18:51:14 Narrator The next video, How to Import NLS Data, demonstrates how to load files from the Investigator download process into STATA, SPSS, R, and SAS.