MANAGING THE DATA
Data should be stored away securely. Any physical data files containing identifying information should be locked away and electronic data files should be password-protected.
Take some time to go through the collected data and identify any questionable responses before you begin data entry, as this will help with making consistent decisions when similar problems arise. Make sure you record any decisions you made during the data entry process such as how you coded the information or which participants were excluded because of too much missing data or unreliable responses. A record of these decisions would allow others to understand and correctly interpret the data if they decide to use your data at a later stage. It would also be helpful if you decide to apply this evaluation approach to another program in the future.
Take some time to go through the collected data and identify any questionable responses before you begin data entry, as this will help with making consistent decisions when similar problems arise. Make sure you record any decisions you made during the data entry process such as how you coded the information or which participants were excluded because of too much missing data or unreliable responses. A record of these decisions would allow others to understand and correctly interpret the data if they decide to use your data at a later stage. It would also be helpful if you decide to apply this evaluation approach to another program in the future.
'Coding' information simply means assigning a number to a categorical response - e.g., Male = 1, Female = 2 or Yes =1, No =2.
An example data record could look like this:
The easiest way to organise quantitative data is to enter it into a spreadsheet. Enter all responses for one participant on the same row , with one data item per column (see example below). Qualitative data, on the other hand, can be entered into a word processing software.
Depending on how much data you have, you may choose to create multiple spreadsheets. For example, if you have three surveys with multiple questions, you may find it confusing to enter all of this information into the same spreadsheet (e.g., there may be more than 90 columns of data). You could create a separate spreadsheet for each of these surveys. When you are ready to analyse your data, you could copy important summary data (most likely total scores, subscores, etc.) to a new spreadsheet that also includes other data you have collected (e.g., age, sex, years of education, etc.)
Before analysing the data, you should double check that all your data has been entered accurately and that your coding is consistent throughout the data file. Some basic steps that you could do include scanning for any values that are out of range (e.g, a response of 7 in a scale with a maximum value of 4 would suggest it is an error) and check what you have coded as missing data are indeed missing.
Before analysing the data, you should double check that all your data has been entered accurately and that your coding is consistent throughout the data file. Some basic steps that you could do include scanning for any values that are out of range (e.g, a response of 7 in a scale with a maximum value of 4 would suggest it is an error) and check what you have coded as missing data are indeed missing.