There may be sets of questions that could be combined into a metric beyond the factor scales already identified (Appendix B). Add to the list of new variables created for analysis purposes, including dummy variables (Appendix A). Data Cleaningħ The SACS data cleaning procedure 3. age) to look for outliers and nonsense values Use frequency tables & histograms to examine the normality of distributions, the range, and extreme skew and kurtosis. Scatterplots for write-in in continuous variables (e.g. Histograms for Likert/ordinal/continuous variables 3.
Frequency tables for categorical variables 2. (e.g., Do the max and min values fall within the question s s expected range? Does the mean make sense for that question?) 1. Perform descriptive statistics to see if the data make sense. Check for and delete duplicate data entries (use SPSS Identify Duplicate Cases procedure or Data Preparation module).
Modify data entry process to reduce future errors Data CleaningĦ The SACS data cleaning procedure 1. Search for and identify error instances 3. 1 Data Cleaning and Missing Data Analysis Dan Merson India McHale April 13, 2010Ģ Overview Introduction to SACS What do we mean by Data Cleaning and why do we do it? The SACS data cleaning procedure The importance of addressing missing data Types of missingness Why we care Missing data analysis procedure What do we do when data are missing? An introduction to weightingģ Student-Athlete Climate Study National, anonymous, population survey of student-athletes' perceptions and experiences pertaining to campus climate and its impact on student-athletes athletes : Academic success Athletic success Athletic identity Almost 9,000 student-athletes from all three divisions and all NCAA sports Focuses on the voices of those who have traditionally not been heard SACSĤ What do we mean by Data Cleaning and why do we do it? Find and eliminate data entry and other errors Examine missing data and perhaps account for it in some way (statistically) Prepare the data for analysis Data Cleaningĥ General data cleaning process 1.
0 kommentar(er)
