Data
Analysis Software for Educators
 |
 |
 |
 |
 |
 |
 |
 |
 |
Creating New Variables Using the New Variable Function The NEW VARIABLE function in EZAnalyze allows you to easily create new variables from your existing data - you can create several different kinds of variables with the NEW VARIABLE function. To create a new variable with your data, select the "New Variable . . ." option from the EZAnalyze menu in Excel. Then choose one of the following: Summary Variable. Select this option if you would like to create a new variable that is the sum or mean of several variables. Common reasons to create a new summary variable are to create an average GPA from students' English, math, social studies, and science GPA's, or to obtain the total number of days a student was absent by adding up the number of absences for each academic quarter. Difference Variable. Select this option if you would like to create a new variable that is a difference score - simply put, one variable subtracted from another variable. Difference variables are useful for showing changes over time - for example, if you started a new program designed to increase attendance in your school, you could create a difference score to show how effective the program was by subtracting the attendance rate after the program (a posttest) from the attendance rate before the program (pretest). Percent Change. This function is useful for determining the difference between two variables in terms of the percent of change from baseline - for example, the amount of change that occurred between a pretest and posttest. This option is similar to the Difference Variable function, except that it provides a more standardized way of reporting the difference between the two variables. Standarized (Z) Score. Select this option if you would like to convert your data to 'standardized scores', or Z scores. Standardizing your variables is useful for putting things 'on the same metric'. For example, if you have two variables, and one is scored on a 5 point scale and the other is scored on a 7 point scale, they are difficult to compare side by side. If you standardize both variables, you can compare the standardized scores side by side easily. You can choose to create standardized scores based on a mean and standard deviation from your own data, or if the population parameters are known, you may choose to use those. Percentile Rank. With this function, you can convert your data to their percentile rank equivalent. For example, if you want to know who is in the top 10% of the senior class at your high school, you can convert their overall GPA into a percentile rank variable to help you see who is in the 90th percentile or higher. Binary Variable. This function creates new variables that are scored as either a 0 or a 1 - a process also known as 'dummy coding'. This is a very useful, and probably underutilized tool. For example, if you wanted to create a disaggregation graph using the percent of people who 'agreed' or 'strongly agreed' with your survey question, you can create a new binary variable that is scored a 1 if people selected 'agree' or 'strongly agree', and a 0 if they did not. This is useful because binary variables that are scored as a 0 or a 1 have a 'special property', and that special property is that the mean of a binary variable with values of 0 or 1 is the percent of people who scored a one. Random Numbers. Using this function, you can quickly create random numbers to demonstrate various statistical problems and concepts. You can set the mean and standard deviation, or generate completely random numbers within a specified range. |
