QUICK GUIDE 3 - MAIN STEPS QUICK GUIDE 3 - MAIN STEPS Check, clean and process raw data (only for quantitative data collection methods) Checking raw data for consistency and completeness is a crucial step in the statistical production process in order to identify potential problems, errors and discrepancies such as outliers, item non-response and miscoding. Data should both be checked against predefined quality and consistency rules, and iteratively by analytically exploring the data set. This can already partly be done during the data collection phase, and it can still be informed by the data analyst’s assessment of the quality of the data set during the analysis phase. This process step can also involve adding data from other sources such as geospatial data sets. Classifying and re-coding existing variables and creating additional variables are further parts of the preparation of the data set before the main analysis. After collecting data, it will be necessary to undertake detailed data analysis, a process where quality control and supervision is crucial to ensure reliable and accurate results. Again, ensuring involvement of a person with advanced training in quantitative data analysis techniques will be necessary for this step. As a general guide, the following stages of data analysis will need to be considered: 1 Data coding/entry Questionnaire data needs to be transformed into another format that is compatible with computer software. As an alternative, mobile data collection through the use of mobile phones/ tablets is becoming increasingly common and can incorporate simple consistency and quality checks at the data collection stage. 28 2 3 Data editing Quality and validation checks or “cleaning” will need to be undertaken so that errors can be found and removed from the data. Data analysis This is the process of analyzing and modelling data in order to highlight useful information and suggest conclusions. Advanced statistical software (such as R or SPSS) may be needed. U N H C R > R E S E A R C H I N G S TAT E L E S S N E S S - Q U I C K G U I D E S 4 Cross tabulation Cross tabulation of census or simple random survey data related to citizenship and statelessness can provide information on the estimated number of stateless persons or persons at risk of statelessness. Through cross tabulation, it may be possible to detect groups within the general population who possess characteristics which in combination serve as strong indicators of statelessness. 5 Data evaluation For high-profile sample surveys, consider undertaking a post-enumeration survey (PES) no more than one month after data collection activity is completed. U N H C R > R E S E A R C H I N G S TAT E L E S S N E S S - Q U I C K G U I D E S 6 Data archiving The data will need to be kept and stored for a set period of time for use over the longer term if necessary or if there’s aneed to refer back to it, perhaps for advocacy purposes. 29