Transcript code number
Coding the Data: Creating Codebooks Chapter 19 Stages in the Research Process Formulate Problem Determine Research Design Determine Data Collection Method Design Data Collection Forms Design Sample and Collect Data Analyze and Interpret the Data Prepare the Research Report Coding • The process of transforming raw data into symbols (usually numbers) that can be utilized for analysis. Example of Likert Scale Strongly Disagree Disagree The celebrity endorser is trustworthy. The celebrity endorser is unattractive. The celebrity endorser is an expert on the product. The celebrity endorser is not knowledgeable about the product. Neither Agree nor Disagree Agree Strongly Agree Coding Likert Scales 1. Give each statement a name: Trustworthy Unattractive Expert Knowledge 2. Assign numbers to each response: 1 = Strongly disagree 2 = Disagree 3 = Neither Agree nor Disagree 4 = Agree 5 = Strongly Agree Example of Semantic-Differential Scale SALESPERSON Not Helpful : : : : : : Helpful Not Friendly : : : : : : Friendly Pestered me : : : : : : Knowledgeable : : : : : : Did Not Pester me Not Knowledgeable Coding Semantic Differential Scales 1. Name each set of bipolar adjectives Helpful Friendly Pester Knowledge 2. Assign numbers to each blank Example for “Helpful” 1 = not helpful 2= 3= 4= Do this for each set. 5= Beware of Reverse Coded Items 6= 7 = helpful Coding Closed-ended Items What is your overall opinion of SEARS department stores? unfavorable Typical coding: favorable 1=unfavorable 2= 3= 4= 5= 6= 7 =favorable Coding Closed-ended Items: Check All That Apply Typical coding: 6 different variables How did you learn about Brown Furniture Company? (check all that apply) (1 if checked; 0 if not) newspaper advertising (1 if checked; 0 if not) radio advertising (1 if checked; 0 if not) billboard advertising (1 if checked; 0 if not) recommended by others (1 if checked; 0 if not) drove by store (1 if checked; 0 if not) other: _______________ Coding Open-ended Items Open-ended items seeking concrete, or factual, responses are relatively easy to code: numeric answers are typically recorded as given by the respondent, while other types of responses are given a specific code number. (1) In what year were you born? (code year) (2) How many times have you eaten at Streeter’s Grill in the last month? (code number) (3) Name the first 3 coffee shops located in Jackson that come to mind. (code as 3 separate variables; assign numbers to represent each coffee shop mentioned) Coding Open-ended Items Open-ended items seeking less structured responses are much more difficult to code. In your own words, give us two or three reasons why you prefer to leave Mississippi after graduation. Process for Coding (Abstract) Open-ended Questions 1. 2. 3. 4. 5. 6. 7. Develop initial response categories (before reading responses) Identify usable responses Review responses; add, delete, revise categories Sort responses into categories, using multiple coders; compare results Repeat #3 and #4 if one or more categories are too broad Assign code numbers for each category; use these codes to represent responses in the data file Assess interrater reliability (the degree of agreement between coders); low interrater reliability suggests that the categories are not well-defined, and #3-6 should be repeated Developing a Codebook SPORTING GOODS SURVEY Please answer the following questions about buying sporting goods over the internet: 1. During the past year, what percentage of the sporting goods you purchased were ordered through the internet? ________ percent 2. How willing are you to purchase merchandise offered through the Avery Sporting Goods web site? Not at all willing 3. Somewhat willing Very willing Please provide some reasons why someone might not want to purchase sporting goods over the internet: Developing a Codebook Avery Sporting Goods – CODEBOOK (partial) Var. Name Description ID questionnaire identification number PERCENT % products purchased through internet (record response) WILLING willingness to purchase through web site 1=not at all willing 2=somewhat willing 3=very willing REASON1 first reason for not purchasing over internet 1=security issues (open ended) 2=no internet access 3=can’t examine goods 4=difficult to return 5=don’t want to wait 6=prior bad exper. w/internet 7=other REASON2 second reason SAME REASON3 third reason SAME What do the data look like? ID PERCENT WILLING REASON1 REASON2 REASON3 1 50 2 4 3 2 0 1 1 6 7 3 20 2 3 4 5 4 90 3 5 5 80 3 5 7