6/24/2016 Marketing Research 1
6/24/2016Marketing Research2 DATA ANALYSIS DATA ENTRY ERROR CHECKING AND VERIFICATION CODING EDITING
The process of checking and adjusting the data ◦ for omissions ◦ for legibility ◦ for consistency And readying them for coding and storage 6/24/2016Marketing Research3
The process of identifying and assigning a numerical score or other character symbol to previously edited data 6/24/2016Marketing Research4
Categories should be exhaustive Categories should be mutually exclusive and independent Remember level of measurement when analyzing data! ◦ Mean for gender is meaningless. Select logical codes ◦ E.g. 1 for most common category ◦ E.g. Low or negative numbers for disagree 6/24/2016Marketing Research5
6/24/2016Marketing Research6 1a. How many years have you been playing tennis on a regular basis (if at all)? Number of years: __________ b. What is your level of play? Novice or non-player... 1Advanced Lower Intermediate.....2Expert Upper Intermediate Teaching Pro c. In the last 12 months, has your level of play improved, remained the same or decreased? Improved Decreased Remained the same Pre-coding Example
May have pre-set categories ◦ Based on your theory/lit review/expectations May develop categories based on responses ◦ Read all responses ◦ Record in abbreviated form ◦ Develop categories ◦ Code a subset to assess categories Do all responses have a clear place? 6/24/2016Marketing Research7
Number each statement ◦ Must determine what makes a “statement” ◦ Usually each complete idea or concept ◦ May or may not follow sentence structure Assign a code to each numbered statement May require several data entry columns 6/24/2016Marketing Research8
Identifies each variable Provides a variable’s description Identifies each code name and position on storage medium Indicates response options and their corresponding codes 6/24/2016Marketing Research9
Add an ID number to each questionnaire ◦ Create category headers at top of each column ◦ Enter ID as the first field ◦ Enter each person across 6/24/2016Marketing Research10
6/24/2016Marketing Research11
Checks: ◦ Look at frequency distributions number of responses how many missing values? consistent with the response categories? make sure there are no “unreasonable” answers If included data not in the range (e.g., “8”), the data needs to be edited Find the questionnaire (with ID) replace with “correct” value includes the full range? e.g, all the numbers in a 1 to 7 scale 6/24/2016Marketing Research12
6/24/2016 Marketing Research 13