Chapter Twelve Data Processing Fundamental Data Analysis and

Chapter Twelve Data Processing, Fundamental Data Analysis, and Statistical Testing of Differences Copyright © 2006 John Wiley & Sons, Inc.

Learning Objectives 1. To develop an understanding of the importance and nature of quality control checks. 2. To understand the data entry process and data entry alternatives. 3. To learn how surveys are tabulated and crosstabulated. 4. To understand the concept of hypothesis development and how to test hypotheses John Wiley & Son, Inc. 2

The Data Analysis Procedure To develop an understanding of the importance and nature of quality control checks • Five Step Procedure for Data Analysis – Step One: Validation and editing (quality control) – Step Two: Coding – Step Three: Data Entry – Step Four: Machine Cleaning of Data – Step Five: Tabulation and Statistical Analysis John Wiley & Son, Inc. 3

Step One: Validation and Editing To develop an understanding of the importance and nature of quality control checks • Validation – The process of ascertaining that interviews actually were conducted as specified. – Telephone Validation • • Was the person actually interviewed? Was the respondent actually qualified? Was the interview conducted in the required manner? Did the interviewer cover the entire survey? – Check for other types of problems – Purpose of the Validation John Wiley & Son, Inc. 4

Step One: Validation and Editing • • • To develop an understanding of the importance and nature of quality control checks Editing Checking for interviewer and respondent mistakes Editing Process 1. Did the interviewer ask or record answers for certain questions? 2. Questionnaires are checked to make sure Skip patterns are followed. 3. Responses to open-ended responses are checked. John Wiley & Son, Inc. 5

Step Two: Coding • To develop an understanding of the importance and nature of quality control checks Coding – Grouping and assigning numeric codes to the responses • The Coding Process 1. 2. 3. 4. Listing responses Consolidating responses Setting codes Entering codes John Wiley & Son, Inc. 6

Step Three: Data Entry To understand the data-entry process and data-entry alternatives. • Data Entry – Process of converting information to a form that can be read by a computer • Intelligent Data Entry – The checking of information being entered for internal logic by either that data entry device or another device connected to it. • The Data Entry Process – The mechanics of the process. – The validated, edited, and coded questionnaires are given to a data entry operator. – The process of going directly from the questionnaire to the data entry device and storage medium is more accurate and efficient. John Wiley & Son, Inc. 7

Step Three: Data Entry To understand the data-entry process and data-entry alternatives. • Scanning – Optical Scanning – Electronically Captured Data is Increasing • • Computer-assisted telephone interviewing Internet surveys Disks-by-mail surveys Touch. Screen Kiosk surveys John Wiley & Son, Inc. 8

Step Four: Machine Cleaning of Data To understand the data-entry process and data-entry alternatives. • Machine Cleaning of Data – A final computerized error check of data. • Error Checking Routines – Check for logical errors in the data • Marginal Report – A computer-generated table of the frequencies of the responses to each question to monitor entry of valid codes and correct use of skip patterns. • Final Error Check in the Process – Should be ready for tabulation and statistical analysis John Wiley & Son, Inc. 9

Step Five: Tabulation and Statistical Analysis • One Way Frequency Tables – – • To learn how surveys are tabulated and cross-tabulated A table showing the number of responses to each answer. The first summary of survey results Options for Base of the Percentages 1. Total respondents 2. Number of people asked the question 3. Number of people answering the question • • Selecting the Base for One-Way Frequency Tables Showing Results from Multiple-Choice Questions John Wiley & Son, Inc. 10

Step Five: Tabulation and Statistical Analysis To learn how surveys are tabulated and cross-tabulated • Cross-Tabulations – Examination of the responses of one question relative to responses to one or more other questions. – Three different percentages calculated for each cell in a crosstabulation table • Column percentage • Row percentage • Total percentages John Wiley & Son, Inc. 11

Graphic Representations of Data • Line Charts – The simplest form of graphs. • Pie Charts – Appropriate for displaying marketing research results in a wide range of situations. • Bar Charts 1. Plain bar chart 2. Clustered bar charts 3. Stacked bar charts 4. Multiple row, three-dimensional bar charts Examples follow slides 13 -19 John Wiley & Son, Inc. 12

Exhibit 12. 11 John Wiley & Son, Inc. Line Chart 13

Exhibit 12. 12 John Wiley & Son, Inc. Pie Chart 14

Exhibit 12. 13 John Wiley & Son, Inc. Simple Two Dimensional Bar Cart 15

Exhibit 12. 14 John Wiley & Son, Inc. Simple Three Dimensional Bar Chart 16

Exhibit 12. 15 John Wiley & Son, Inc. Clustered Bar Chart 17

Exhibit 12. 16 John Wiley & Son, Inc. Stacked Bar Chart 18

Exhibit 12. 17 John Wiley & Son, Inc. Multiple-Row, Three Dimensional Bar Chart 19

Descriptive Statistics • Measures of Central Tendency – Nominal and Ordinal Scales – Interval and Ratio Scales – Mean – Median – Mode John Wiley & Son, Inc. 20

Descriptive Statistics Measures of Central Tendency • Formula for the Mean h X where = I=1 f i. X i n fi = the frequency of the ith class Xi = the midpoint of that class h = the number of classes n = the total number of observations John Wiley & Son, Inc. 21

Descriptive Statistics • Measures of Dispersion – Standard deviation – Variance • The sums of the squared deviations from the mean divided by the number of observations minus one. • The same formula as standard deviation with the squareroot sign removed. – Range • The maximum value for a variable minus the minimum value for that variable John Wiley & Son, Inc. 22

Descriptive Statistics Measures of Dispersion Standard deviation S = where √ n I=1 (Xi - X) 2 n-1 S = sample standard deviation Xi = the value of the ith observation X = the sample mean n = the sample size John Wiley & Son, Inc. 23

Descriptive Statistics • Percentages, and Statistical Tests – Whether to use measures of central tendency or percentages. – Responses are either categorical or take the form of continuous variables • Variables such as age can be continuous or categorical. • If categories are used, one-way frequency tables and crosstabulations are used for analysis – Continuous data can be put into categories. • Evaluating Differences and Changes John Wiley & Son, Inc. 24

Statistical Significance • Statistical Inference – To generalize from sample results to population characteristics • Three Concepts of Differences – Mathematical differences – Statistical significance – Managerially important differences John Wiley & Son, Inc. 25

Hypothesis Testing To understand the concept of hypothesis development and how to test hypotheses. • Hypothesis – An assumption that a researcher makes about some characteristic of the population under study. • Explanation for Differences between a Hypothesized Value and a Particular Research Result – The Hypothesis is true and the observed difference is likely due to sampling error – The Hypothesis is false and the true value is some other value John Wiley & Son, Inc. 26

Hypothesis Testing To understand the concept of hypothesis development and how to test hypotheses. • Steps in Hypothesis Testing – Step One: Stating the Hypothesis • Null hypothesis: Ho • Alternative hypothesis: Ha – Step Two: Choosing the Appropriate Test Statistic • Exhibit 12. 20 Statistical Tests and Their Uses—provides a guide to selecting the appropriate test for various situations John Wiley & Son, Inc. 27

Hypothesis Testing To understand the concept of hypothesis development and how to test hypotheses. – Step Three: Developing a Decision Rule • Significance level (α)— 0. 01, 0. 05, or 0. 10—that will determine whether to reject or fail to reject the null hypothesis – Step Four: Calculating the Value of the Test Statistic • Use the appropriate formula • Compare calculated value to the critical value. • State the result in terms of: – rejecting the null hypothesis – failing to reject the null hypothesis – Step Five: Stating the Conclusion • Summarizes the results of the test—should be stated from the perspective of the original research question John Wiley & Son, Inc. 28

Hypothesis Testing To understand the concept of hypothesis development and how to test hypotheses. • Types of Errors in Hypothesis Testing – Type I Error • Rejection of the null hypothesis when, in fact, it is true. 1 – α is the probability of making a correct decision by not rejecting the null hypothesis when, in fact, it is true – Type II Error • Acceptance of the null hypothesis when, in fact, it is false. 1– β reflects the probability of making a correct decision in rejecting the null hypothesis when, in fact, it is false – Accepting Ho or Failing to Reject Ho? • Is there enough data to conclude that Ho is correct • One-Tailed Test or Two-Tailed Test? John Wiley & Son, Inc. 29

Table 12. 21 Type I and Type II Errors Actual State of the Null Hypothesis Fail to Reject Ho Ho is true Correct (1 - ) no error Type I error ( ) Ho is false Type II error ( ) Correct (1 - ) no error John Wiley & Son, Inc. 30

Commonly Used Statistical Hypothesis Tests To understand the concept of hypothesis development and how to test hypotheses. • Independent Versus Related Samples – Independent samples • Measurement of a variable in one population has no effect on the measurement of the other variable – Related Samples • Measurement of a variable in one population may influence the measurement of the other variable. • Degrees of Freedom – The number of observations minus the number of constraints. – The number of degrees of freedom John Wiley & Son, Inc. 31

Commonly Used Statistical Hypothesis Tests • To understand the concept of hypothesis development and how to test hypotheses. p - VALUES AND SIGNIFICANCE TESTING – P- Value • The exact probability of getting a computed test statistic that was largely due to chance 1. The smaller the p-value, the smaller the probability that the observed result occurred by chance. 2. The p-value is the demanding level of statistical significance that can be met, based on the calculated value of the statistic John Wiley & Son, Inc. 32

SUMMARY • • Overview of the Data Analysis Procedure Step One: Validation and Editing Step Two: Coding Step Three: Data Entry Step Four: Machine Cleaning of Data Step Five: Tabulation and Statistical Analysis Graphic Representations of Data Descriptive Statistics John Wiley & Son, Inc. 33

SUMMARY • Statistical Significance • Hypothesis Testing • Commonly Used Statistical Hypothesis Tests • P-Values and Significance Testing • Statistics on the Internet John Wiley & Son, Inc. 34

The End Copyright © 2006 John Wiley & Son, Inc. 35