MKT 6309 Exam 2

Organizing Your Written Report

There are certain elements that must be considered when you are preparing the report. • These elements can be grouped in three sections: • Front section • Main section • End section

Estimating Variability

There are three ways to estimate variance p times q: 1. Unless we have other information we can assume the most conservative case and use the maximum amount of variance we would expect in the population (p= 50%, q= 50%) 2. Use data from a previous study conducted on the same population 3. Conduct a small pilot study to estimate variance

Square of the part correlation coefficient

This coefficient represents an increase in R2 when a variable is entered into a regression equation that already contains the other independent variables

Square of the partial correlation coefficient.

This measure, R2 yxi.xjxk, is the coefficient of determination between the dependent variable and the independent variable, controlling for the effects of the other independent variables.

Square of the simple correlation coefficient

This measure, r2, represents the proportion of the variation in the dependent variable explained by the independent variable in a bivariate relationship.

Nonparametric tests

assume that the variables are measured on a nominal or ordinal scale. These tests can be further classified based on whether one or two or more samples are involved.

Parametric tests

assume that the variables of interest are measured on at least an interval scale. These tests can be further classified based on whether one or two or more samples are involved.

Analysis of variance (ANOVA)

) is used as a test of means for two or more populations. The null hypothesis, typically, is that all means are equal. • Analysis of variance must have a dependent variable that is metric (measured using an interval or ratio scale). • There must also be one or more independent variables that are all categorical (nonmetric). Categorical independent variables are also called factors.

Parameter

- A characteristic or measure of a population - If it were possible to take measures from all members of a population without error, a true value of a parameter could be determined

Statistic

- A characteristic or measure of a sample - Statistics are calculated from sample data and used to estimate population parameters

Discard Unsatisfactory Respondents

- In this approach, the respondents with unsatisfactory responses are simply discarded.

Forward inclusion

. Initially, there are no predictor variables in the regression equation. Predictor variables are entered one at a time, only if they meet certain criteria specified in terms of F ratio. The order in which the variables are included is based on the contribution to the explained variance.

Standard error of estimate.

. This statistic, SEE, is the standard deviation of the actual Y values from the predicted Ŷ values.

Codebook

Contains coding instructions and the necessary information about variables in the data set. A codebook generally contains the following information: • column number • record number • variable number • variable name • question number • instructions for coding

Procedures for Drawing Simple Random Samples

1. Select a suitable sampling frame. 2. Assign each element a number from 1 to N (population size). 3. Generate n (sample size) different random numbers between 1 and N. This can be done using a software package or using a table of simple random numbers. 4. The numbers generated denote the elements that should be included in the sample. Members selected should be representative of all the members of the population.

Story telling with data by Cole Nussbaumer Knaflic

1. Understand the context • Exploratory vs Explanatory • Who • What • How • presentation vs report • time available 2. Choose an effective visual 3. Eliminate clutter 4. Focus your audience's attention 5. Think like a designer 6. Tell a story • Questions to consider in telling a story (Cliff Atkinson, Beyond Bullet points) • The setting (when and where does it take place) • The main character (who is driving the action?) • The imbalance (why is it necessary, what has changed) • The balance (what do you want to see happen?) • The solution (how will you bring about the change?)

Convenience Samples

A convenience sample is drawn at the convenience of the researcher or interviewer • Often, respondents are selected because they happen to be in the right place at the right time. • Use of students, and members of social organizations • Mall intercept interviews without qualifying the respondents • Department stores using charge account lists • "People on the street" interviews

Accuracy

A convenient way to describe the amount of sample error due to the size of the sample, or the accuracy of a sample, is to treat it as a plus-or-minus percentage value. • The accuracy of a sample is usually expressed as a ±% such as ±5% or ±1%.

Strength of Association

The strength of association is measured by the square of the multiple correlation coefficient, R2, which is also called the coefficient of multiple determination.

Illustrative Applications of One-Way ANOVA

A department store is attempting to determine the effect of in-store promotion (X) on sales (Y). For the purpose of illustrating hand calculations, the data of Table 16.2 are transformed in Table 16.3 to show the store sales (Yij) for each level of promotion

Referral or Snowball Sampling

A form of judgement sampling • After being interviewed, respondents are asked to identify others who belong to the target population of interest. • Subsequent respondents are selected based on the referrals. • Referral samples are most appropriate when there is a limited sample frame and when respondents can provide the names of others who would qualify for the survey • Members of the population who are less well known, disliked, or whose opinions conflict with the respondent have a low probability of being selected into a referral sample.

Kruskal-Wallis one-way analysis of variance.

A more powerful test is the Kruskal-Wallis one-way analysis of variance. This is an extension of the Mann-Whitney test (Chapter 15). This test also examines the difference in medians. All cases from the k groups are ordered in a single ranking. If the k populations are the same, the groups should be similar in terms of ranks within each group. The rank sum is calculated for each group. From these, the Kruskal-Wallis H statistic, which has a chisquare distribution, is computed. • The Kruskal-Wallis test is more powerful than the k-sample median test as it uses the rank value of each case, not merely its location relative to the median. However, if there are a large number of tied rankings in the data, the k-sample median test may be a better choice.

Substitute a Neutral Value -

A neutral value (typically the mean response to the variable) is substituted for the missing responses.

partial correlation coefficient

A partial correlation coefficient measures the association between two variables after controlling for, or adjusting for, the effects of one or more additional variables. • For example: association between attitude towards a city (y) and duration of residence (x1) after controlling for a third variable, importance attached to weather (x2) • Partial correlations have an order associated with them. The order indicates how many variables are being adjusted or controlled. • The simple correlation coefficient, r, has a zero-order, as it does not control for any additional variables while measuring the association between two variables. • The coefficient rxy.z is a first-order partial correlation coefficient, as it controls for the effect of one additional variable, Z. • A second-order partial correlation coefficient controls for the effects of two variables, a third-order for the effects of three variables, and so on. • The special case when a partial correlation is larger than its respective zero-order correlation involves a suppressor effect

treatment.

A particular combination of factor levels, or categories, is called a

Questionnaire Checking

A questionnaire returned from the field may be unacceptable for several reasons. • Parts of the questionnaire may be incomplete. • The pattern of responses may indicate that the respondent did not understand or follow the instructions. • The responses show little variance. • One or more pages are missing. • The questionnaire is received after the cutoff date. • The questionnaire is answered by someone who does not qualify for participation.

Examination of Residuals

A residual is the difference between the observed value of Yi and the value predicted by the regression equation Ŷi . • Scattergrams of the residuals, in which the residuals are plotted against the predicted values, Ŷi, time, or predictor variables, provide useful insights in examining the appropriateness of the underlying assumptions and regression model fit. • The assumption of a normally distributed error term can be examined by constructing a histogram of the residuals. • The assumption of constant variance of the error term can be examined by plotting the residuals against the predicted values of the dependent variable, Ŷi . • A plot of residuals against time, or the sequence of observations, will throw some light on the assumption that the error terms are uncorrelated. • Plotting the residuals against the independent variables provides evidence of the appropriateness or inappropriateness of using a linear model. Again, the plot should result in a random pattern. • To examine whether any additional variables should be included in the regression equation, one could run a regression of the residuals on the proposed variables. • If an examination of the residuals indicates that the assumptions underlying linear regression are not met, the researcher can transform the variables in an attempt to satisfy the assumptions.

Scattergram

A scatter diagram, or scattergram, is a plot of the values of two variables for all the cases or observations. The most commonly used technique for fitting a straight line to a scattergram is the least-squares procedure. In fitting the line, the least-squares procedure minimizes the sum of squared errors

t statistic

A t statistic with n − 2 degrees of freedom can be used to test the null hypothesis that no linear relationship exists between X and Y, or H0: β = 0, where t = b /SEb

Guidelines on Supervision

All research projects should be properly supervised. It is the data collection agency's responsibility to: • Properly supervise interviews. • Monitor an agreed-upon proportion of interviewers' telephone calls. • Be available to report on the status of the project daily to the project director, unless otherwise instructed. • Keep all studies, materials, and findings confidential. • Notify concerned parties if the anticipated schedule is not met. • Attend all interviewer briefings. • Keep current and accurate records of the interviewing progress. • Make sure all interviewers have required materials on time. • Edit each questionnaire. • Provide consistent and positive feedback to interviewers. • Not falsify any work.

Standardized regression coefficient

Also termed the beta coefficient or beta weight, this is the slope obtained by the regression of Y on X when the data are standardized.

Amount of Diversity or Variation

As diversity/variation increases, larger samples are required

Degree of Precision

As need for precision increases, larger samples are required

Coding

Assign a code, usually a number, to each possible response to each question. The code includes an indication of the column position (field) and data record it will occupy. Coding Questions • Ensure fixed field codes, which means that the number of records for each respondent is the same and the same data appear in the same column(s) for all respondents, are highly desirable. • Coding of structured questions is relatively simple, since the response options are predetermined. • In questions that permit a large number of responses, each possible response option should be assigned a separate column.

Hypothesis Testing Step Three

Choose a Level of Significance Type I Error • Type I error occurs when the sample results lead to the rejection of the null hypothesis when it is in fact true. • The probability of type I error (α) is also called the level of significance. • Typically a value of 0.05 is selected for α Type II Error • Type II error occurs when, based on the sample results, the null hypothesis is not rejected when it is in fact false. • The probability of type II error is denoted by β. • Unlike α, which is specified by the researcher, the magnitude of β depends on the actual value of the population parameter (mean or proportion).

Hypothesis Testing Step Four

Collect Data and Calculate Test Statistic • If 30 customers were surveyed and 17 shopped online, the value of the sample proportion is p = 17/30 = 0.567

Hypothesis Testing Step Six and Seven

Compare the Probability (Critical Value) & Making the Decision • If the probability associated with the calculated or observed value of the test statistic (TSCAL) is less than the level of significance (α), the null hypothesis is rejected. • The probability associated with the calculated or observed value of the test statistic is 0.0301 (this is the probability of getting a p value of 0.567 when π = 0.40). • This is less than the level of significance of 0.05. Hence, the null hypothesis is rejected.

Degree of Confidence

Confidence increases as sample size increases

Data Cleaning Consistency Checks

Consistency checks identify data that are out of range, logically inconsistent, or have extreme values. • Computer packages like SPSS, SAS, and EXCEL can be programmed to identify out-of-range values for each variable and print out the respondent code, variable code, variable name, record number, column number, and out-ofrange value. • Extreme values should be closely examined.

Hypothesis Testing Step One

Formulate the Hypothesis • A null hypothesis may be rejected, but it can never be accepted based on a single test. In classical hypothesis testing, there is no way to determine whether the null hypothesis is true. • In marketing research, the null hypothesis is formulated in such a way that its rejection leads to the acceptance of the desired conclusion. The alternative hypothesis represents the conclusion for which evidence is sought. - Example = introduction of online shopping service if the proportion of online users is greater than 40% • In the previous case, the test of the null hypothesis is a onetailed test, because the alternative hypothesis is expressed directionally. • If the researcher wanted to determine whether the proportion of customers who shop on the internet is different from 40%, then a two-tailed test would be required, and the hypotheses would be expressed as: p is equal to .40 or p is not equal to .40

Stepwise solution

Forward inclusion is combined with the removal of predictors that no longer meet the specified criterion at each step.

Assign Missing Values

If returning the questionnaires to the field is not feasible, the editor may assign missing values to unsatisfactory responses

Nonmetric Correlation

If the nonmetric variables are ordinal and numeric, Spearman's rho (ρ), and Kendall's tau (τ) are two measures of nonmetric correlation, which can be used to examine the correlation between them. • Both these measures use rankings rather than the absolute values of the variables, and the basic concepts underlying them are quite similar. Both vary from −1.0 to +1.0. • In the absence of ties, Spearman's ρs yields a closer approximation to the Pearson product moment correlation coefficient,ρ , than Kendall's τ. In these cases, the absolute magnitude of τ tends to be smaller than Pearson's ρ. • On the other hand, when the data contain a large number of tied ranks, Kendall's τ seems more appropriate.

Issues in Interpretation - Multiple Comparisons

If the null hypothesis of equal means is rejected, we can only conclude that not all of the group means are equal. We may wish to examine differences among specific means. This can be done by specifying appropriate contrasts or comparisons used to determine which of the means are statistically different. • A priori contrasts are determined before conducting the analysis, based on the researcher's theoretical framework. Generally, a priori contrasts are used in lieu of the ANOVA F test. The contrasts selected are orthogonal (they are independent in a statistical sense). A posteriori contrasts are made after the analysis. These are generally multiple comparison tests. They enable the researcher to construct generalized confidence intervals that can be used to make pairwise comparisons of all treatment means. These tests, listed in order of decreasing power, include least significant difference, Duncan's multiple range test, Student-Newman-Keuls, Tukey's alternate procedure, honestly significant difference, modified least significant difference, and Scheffe's test. Of these tests, least significant difference is the most powerful, Scheffe's the most conservative.

Statistical significance

If the partial regression coefficient of a variable is not significant, as determined by an incremental F test, that variable is judged to be unimportant. An exception to this rule is made if there are strong theoretical reasons for believing that the variable is important.

Issues in Interpretation

Important issues involved in the interpretation of ANOVA results include interactions, relative importance of factors, and multiple comparisons. Interactions • The different interactions that can arise when conducting ANOVA on two or more factors are shown in Figure 16.3. Relative Importance of Factors • Experimental designs are usually balanced, in that each cell contains the same number of respondents. This results in an orthogonal design in which the factors are uncorrelated. Hence, it is possible to determine unambiguously the relative importance of each factor in explaining the variation in the dependent variable.

Conducting One-Way ANOVA

In analysis of variance, we estimate two measures of variation: within groups (SSwithin) and between groups (SSbetween). Thus, by comparing the Y variance estimates based on between-group and within-group variation, we can test the null hypothesis.

N-Way ANOVA

In marketing research, one is often concerned with the effect of more than one factor simultaneously. For example: • How do advertising levels (high, medium, and low) interact with price levels (high, medium, and low) to influence a brand's sale? • Do educational levels (less than high school, high school graduate, some college, and college graduate) and age (less than 35, 35-55, more than 55) affect consumption of a brand? • What is the effect of consumers' familiarity with a department store (high, medium, and low) and store image (positive, neutral, and negative) on preference for the store?

Backward elimination

Initially, all the predictor variables are included in the regression equation. Predictors are then removed one at a time based on the F ratio for removal.

Level of Confidence: z

It is customary among marketing researchers to use the 95% level of confidence. - For 95%, z= 1.96 OR - For 99%, z= 2.58 • We use the phrase "level of confidence" because it refers to how confident we are that the sample finding will repeat itself if we conducted a different survey tomorrow or again the next day By setting z= 1.96, it means that if we were to conduct our survey over a 100 times, 95 of these times we would get a sample finding that would fall within our predetermined level of accuracy, e. • This gives us some confidence in the reliability of our sample estimate.

Supervision of Field Workers

Make sure they are following the procedures and techniques in which they were trained. Including: • Quality Control and Editing - This requires checking to see if the field procedures are being properly implemented. • Sampling Control - The supervisor attempts to ensure that the interviewers are strictly following the sampling plan. • Control of Cheating - Cheating can be minimized through proper training, supervision, and validation. • Central Office Control - Supervisors provide quality and cost-control information to the central office.

Hypothesis Testing Step Eight

Marketing Research Conclusion • The conclusion reached by hypothesis testing must be expressed in terms of the marketing research problem. • In our example, we conclude that there is evidence that the proportion of customers who shop online is significantly greater than 0.40. Hence, the recommendation to the department store would be to introduce the new online shopping service.

One-Way Analysis of Variance

Marketing researchers are often interested in examining the differences in the mean values of the dependent variable for several categories of a single independent variable or factor. For example: • Do the various segments differ in terms of their volume of product consumption? • Do the brand evaluations of groups exposed to different commercials vary? • What is the effect of consumers' familiarity with the store (measured as high, medium, and low) on preference for the store?

Nonmetric ANOVA

Nonmetric analysis of variance examines the difference in the central tendencies of more than two groups when the dependent variable is measured on an ordinal scale. • One such procedure is the k-sample median test. As its name implies, this is an extension of the median test for two groups.

Nonprobability Sampling

Nonprobability sampling plans are sometimes used: - First, they are fast, simple to use, and less costly than probability sampling plans - Second, many managers are perfectly happy to ask n number of persons a question to help them make a decision • With a nonprobability sampling method, all members of the population do not have a known probability of being selected into the sample. • Because of this we cannot say that a sample drawn using a nonprobability sampling method is representative of some larger population. There are four nonprobability sampling methods: - Convenience samples - Judgment samples - Referral samples - Quota samples

Repeated Measures ANOVA

One way of controlling the differences between subjects is by observing each subject under each experimental condition (see Table 16.7). Since repeated measurements are obtained from each respondent, this design is referred to as within subjects design or repeated measures analysis of variance. Repeated measures analysis of variance may be thought of as an extension of the paired-samples t test to the case of more than two related samples.

Improving the Efficiency of Report Writing

Online reporting software electronically distributes marketing research reports to selected managers in an interactive format that allows each user to conduct his or her own analyses.

Standard error.

The standard deviation of b, SEb, is called the standard error.

Hypothesis Testing Related to Differences

Parametric tests Nonparametric tests Independent samples Paired Samples

At any given sample size, there is a trade-off between confidence and

Precision. Higher precision means lower confidence unless we can increase the sample size

Adjusted R2

R2 coefficient of multiple determination, is adjusted for the number of independent variables and the sample size to account for the diminishing returns. After the first few variables, the additional independent variables do not make much contribution.

Online Sampling Techniques

Random online intercept sampling relies on a random selection of website visitors • Invitation online sampling is when potential respondents are alerted that they may fill out a questionnaire that is hosted at a specific website • Online panel sampling refers to consumer or other respondent panels that are set up by marketing research companies for the explicit purpose of conducting online surveys with representative samples

Ethical Issues in Field Work

Researchers and field workers should: • make respondents feel comfortable by addressing their apprehensions and concerns. • respect respondents' time, feelings, privacy, and right to self-determination. • follow the accepted procedures for the selection, training, supervision, validation, and evaluation of field workers. • carefully document field work procedures and make them available to clients.

Editing and treatment of unsatisfactory results

Return to the Field Assign Missing Values Discard Unsatisfactory Respondents

Formula to Determine Sample Accuracy

Sample Size Formula When Estimating a Percentage n = z2 (p * q)/ e2 Where: n= sample size z= Standard error associated with chosen level of confidence (1.96 or 2.58) p=estimated % in population q=(100%-p) e= acceptable error (accuracy)

Hypothesis Testing Step Two

Select an Appropriate Test • The test statistic measures how close the sample has come to the null hypothesis. • The test statistic often follows a well-known distribution, such as the normal t, or chi-square distribution. • Generally, z-tests are used when we have large sample sizes (n > 30), whereas t-tests are most helpful with a smaller sample size (n < 30). Both methods assume a normal distribution of the data, but the z-tests are most useful when the standard deviation is known. In our example, the z statistic, which follows the standard normal distribution, would be appropriate.

Kolmogorov-Smirnov (K-S) one-sample test

Sometimes the researcher wants to test whether the observations for a particular variable could reasonably have come from a particular distribution, such as the normal, uniform, or Poisson distribution. The Kolmogorov-Smirnov (K-S) one-sample test is one such goodness-of-fit test. The K-S compares the cumulative distribution function for a variable with a specified distribution. examines whether the two distributions are the same. It takes into account any differences between the two distributions, including the median, dispersion, and skewness.

Data Cleaning Treatment of Missing Responses

Substitute a Neutral Value Substitute an Imputed Response casewise deletion pairwise deletion

Validation of Fieldwork

Supervisors should: • call 10-25% of the respondents to inquire whether the field workers actually conducted the interviews. • ask about the length and quality of the interview, reaction to the interviewer, and basic demographic data. • Cross-check demographic information against the information reported by the interviewers on the questionnaires.

Guidelines on Interviewing

Techniques for good interviewing: • Provide his or her full name, if asked by the respondent, as well as a phone number for the research firm. • Read each question exactly as written. Report any problems to the supervisor as soon as possible. • Read the questions in the order indicated on the questionnaire, following the proper skip sequences. • Clarify any question by the respondent in a neutral way. • Not mislead respondents as to the length of the interview. • Not reveal the ultimate client's identity unless instructed to do so. • Keep a tally on and the reason for each terminated interview. • Remain neutral, do not indicate (dis)agreement with the respondent. • Speak slowly and distinctly. • Record all replies verbatim, not paraphrased. • Avoid unnecessary conversation with the respondent. • Probe and clarify in a neutral manner for additional comments on all open-ended questions, unless otherwise indicated. • Write neatly and legibly. • Check all work for thoroughness before turning in to the supervisor. • When terminating a respondent, do it neutrally. • Keep all studies, materials, and findings confidential. • Not falsify any interviews or any answers to any question. • Thank the respondent for participating in the study

Significance Testing

Testing for the significance of the β′is can be done in a manner similar to that in the bivariate case by using t tests.

F test

The F test is used to test the null hypothesis that the coefficient of multiple determination in the population, R2 pop, is zero. This is equivalent to testing the null hypothesis. The test statistic has an F distribution with k and (n − k − 1) degrees of freedom.

SAS Enterprise Guide

The Summary Statistics task provides summary statistics including basic summary statistics, percentile summary statistics, and more advanced summary statistics including confidence intervals, t-statistics, coefficient of variation, and sums of squares. • The Summary Statistics task also provides graphical displays, including histograms and box-and-whisker plots. • The One-Way Frequencies task can be used to generate frequency tables as well as binominal and chi-square tests.

Sum of squared errors

The distances of all the points from the regression line are squared and added together to arrive at the sum of squared errors, which is a measure of total error,

Regression coefficient

The estimated parameter b is usually referred to as the non-standardized regression coefficient.

Partial F test

The significance of a partial regression coefficient,βi , of Xi may be tested using an incremental F statistic. The incremental F statistic is based on the increment in the explained sum of squares resulting from the addition of the independent variable Xi to the regression equation after all the other independent variables have been included.

Front Section

The front section consists of all pages that precede the first page of the report: • Title page • Letter of authorization/approval of research proposal • Table of contents • List of illustrations • Executive summary

SPSS Windows

The main program in SPSS is FREQUENCIES. It produces a table of frequency counts, percentages, and cumulative percentages for the values of each variable. It gives all of the associated statistics. • If the data are interval scaled and only the summary statistics are desired, the DESCRIPTIVES procedure can be used. • The EXPLORE procedure produces summary statistics and graphical displays, either for all of the cases or separately for groups of cases. Mean, median, variance, standard deviation, minimum, maximum, and range are some of the statistics that can be calculated.

Measures based on standardized coefficients or beta weights

The most commonly used measures are the absolute values of the beta weights, |Bi|, or the squared values, Bi 2.

Stepwise regression

The order in which the predictors enter or are removed from the regression equation is used to infer their relative importance.

Partial regression coefficient

The partial regression coefficient, b1, denotes the change in the predicted value, Ŷ, per unit change in X1 when the other independent variables, X2 to Xk, are held constant

Power of a Test

The power of a test is the probability (1 − β) of rejecting the null hypothesis when it is false and should be rejected. • Although β is unknown, it is related to α. An extremely low value of α (e.g.,α = 0.001) will result in intolerably high β errors. • So it is necessary to balance the two types of errors.

Coefficient of multiple determination

The strength of association in multiple regression is measured by the square of the multiple correlation coefficient, R2, which is also called the coefficient of multiple determination.

Product Moment Correlation

The product moment correlation, r, summarizes the strength of association between two metric (interval or ratio scaled) variables, say X and Y. • It is an index used to determine whether a linear or straightline relationship exists between X and Y. • As it was originally proposed by Karl Pearson, it is also known as the Pearson correlation coefficient, simple correlation, bivariate correlation, or merely the correlation coefficient. The correlation coefficient between two variables will be the same regardless of their underlying units of measurement.

Stepwise Regression

The purpose of stepwise regression is to select, from a large number of predictor variables, a small subset of variables that account for most of the variation in the dependent or criterion variable. In this procedure, the predictor variables enter or are removed from the regression equation one at a time. There are several approaches to stepwise regression. •

Presenting Your Research

The purpose of the oral presentation is to succinctly present the information and to provide an opportunity for questions and discussion. • Prepare carefully for your presentation • Identify and analyze your audience. • Who is the audience? • What message do you wish to communicate? • Are there multiple audiences? • Predetermine expectations of the audience, formal vs informal presentation? Handouts? PPT Presentation? • Present your points clearly and succinctly. • Practice, practice, practice! • Dress appropriately.

Return to the Field

The questionnaires with unsatisfactory responses may be returned to the field, where the interviewers recontact the respondents.

Substitute an Imputed Response -

The respondents' pattern of responses to other questions are used to impute or calculate a suitable response to the missing questions.

Assumptions in ANOVA

The salient assumptions in analysis of variance can be summarized as follows: 1. Ordinarily, the categories of the independent variable are assumed to be fixed. Inferences are made only to the specific categories considered. This is referred to as the fixed-effects model. 2. The error term is normally distributed, with a zero mean and a constant variance. The error is not related to any of the categories of X. 3. The error terms are uncorrelated. If the error terms are correlated (i.e., the observations are not independent), the F ratio can be seriously distorted.

Guidelines on Interviewer Training

Training should cover the following: • The research process: how a study is developed, implemented, and reported. • Importance of: interviewer's honesty, objectivity, and professionalism. • Confidentiality of the respondent and client. • Familiarity with market research terminology. • Importance of following the exact wording and recording responses verbatim. • Purpose and use of probing and clarifying techniques. • The reason for and use of classification & respondent information questions. • A review of samples of instructions and questionnaires. • Importance of the respondent's positive feelings about survey research.

Simple Random Sampling

Used when the population is small and can easily be counted or even when the population is large but is contained in an electronic database which can automatically "draw" a random sample • Probability of selection = sample size/population size • With simple random sampling the probability of being selected into the sample is "known" and equal for all members of the population.

Precision level

When estimating a population parameter by using a sample statistic, the precision level is the desired size of the estimating interval. This is the maximum permissible difference between the sample statistic and the population parameter

Analysis of Covariance

When examining the differences in the mean values of the dependent variable related to the effect of the controlled independent variables, it is often necessary to take into account the influence of uncontrolled independent variables. For example: • In determining how different groups exposed to different commercials evaluate a brand it may be necessary to control for prior knowledge. • In determining how different price levels will affect a household's cereal consumption it may be essential to take household size into account. We again use the data of Table 16.2 to illustrate analysis of covariance. • Suppose that we wanted to determine the effect of in-store promotion and couponing on sales while controlling for the effect of clientele. The results are shown in Table 16.6.

Sample Size and Accuracy

With small increases in sample size, we can gain increases in sample accuracy up to a point (about 500). Beyond that point, there are diminishing returns in accuracy With small increases in sample size, we can gain increases in sample accuracy up to a point. Beyond that point, there are diminishing returns in accuracy.

Random Digit Dialing (RDD)

a method of randomly generating numbers to represent telephone numbers - This approach is used in telephone surveys to overcome the problems of unlisted and new telephone numbers

sample

a subset of the population that should represent the population

Wilcoxon matched-pairs signed-ranks test

analyzes the differences between paired observations, taking into account the magnitude of the differences. • It computes the differences between the pairs of variables and ranks the absolute differences. • The next step is to sum the positive and negative ranks. • The test statistic, z, is computed from the positive and negative rank sums. - Example - The example considered for the paired t test, whether the respondents differed in terms of attitude toward the Internet and attitude toward technology, is considered again. Suppose we assume that both these variables are measured on ordinal rather than interval scales. Accordingly, we use the Wilcoxon test. The results are shown in Table 15.18.

Sampling error

any error in a survey that occurs because a sample is used. - Usually caused by method of sample selection or sample size

Multicollinearity

arises when intercorrelations among the predictors are very high. Multicollinearity can result in several problems, including: - The partial regression coefficients may not be estimated precisely. The standard errors are likely to be high. - The magnitudes, as well as the signs of the partial regression coefficients, may change from sample to sample. - It becomes difficult to assess the relative importance of the independent variables in explaining the variation in the dependent variable. - Predictor variables may be incorrectly included or removed in stepwise regression. • A simple procedure for adjusting for multicollinearity consists of using only one of the variables in a highly correlated set of variables. • Alternatively, the set of independent variables can be transformed into a new set of predictors that are mutually independent by using techniques such as principal components analysis. • More specialized techniques, such as ridge regression and latent root regression, can also be used.

chi-square test

can also be performed on a single variable from one sample. In this context, the chi-square serves as a goodness-of-fit test.

casewise deletion

cases or respondents with any missing responses, are discarded from the analysis.

two-sample median test

determines whether the two groups are drawn from populations with the same median. It is not as powerful as the Mann-Whitney U test because it merely uses the location of each observation relative to the median, and not the rank, of each observation.

Regression Analysis

examines associative relationships between a metric dependent variable and one or more independent variables in the following ways: • Determine whether the independent variables explain a significant variation in the dependent variable. Whether a relationship exists. • Determine how much of the variation in the dependent variable can be explained by the independent variables: strength of the relationship. • Determine the structure or form of the relationship: the mathematical equation relating the independent and dependent variables. • Predict the values of the dependent variable. • Control for other independent variables when evaluating the contributions of a specific variable or set of variables. • Regression analysis is concerned with the nature and degree of association between variables and does not imply or assume any causality

pairwise deletion

instead of discarding all cases with any missing values, the researcher uses only the cases or respondents with complete responses for the variables involved in each calculation

One-way analysis of variance

involves only one categorical variable, or a single factor. In one-way analysis of variance, a treatment is the same as a factor level.

runs test

is a test of randomness for the dichotomous variables. This test is conducted by determining whether the order or sequence in which observations are obtained is random.

binomial test

is also a goodness-of-fit test for dichotomous variables. It tests the goodness of fit of the observed number of observations in each category to the number expected under a specified binomial distribution.

disproportionate sample size

is any other allocation that would occur if we based our sample size per stratum not on its proportionate share of the population but on its variance as per our sample size formula. We should use a disproportionate sample size allocation in order to achieve the statistical efficiency possible by using a stratified sample plan. • In other words, we would allocate more sample size to the strata with the higher variances and take sample size away from strata with lower variances. • This process gives us "statistical efficiency." We are able to estimate population facts for each stratum accurately without increasing the total sample size required to do the study In a stratified sample, we would be estimating sample findings per stratum.

sign test

is not as powerful as the Wilcoxon matched pairs signed-ranks test as it only compares the signs of the differences between pairs of variables without taking into account the ranks.

Multivariate analysis of variance (MANOVA)

is similar to analysis of variance (ANOVA), except that instead of one metric dependent variable, we have two or more. • In MANOVA, the null hypothesis is that the vectors of means on multiple dependent variables are equal across groups. • Multivariate analysis of variance is appropriate when there are two or more dependent variables that are correlated.

Standardization

is the process by which the raw data are transformed into new variables that have a mean of 0 and a variance of 1. • When the data are standardized, the intercept assumes a value of 0.

beta coefficient or beta weight

is used to denote the standardized regression coefficient

Plus-one dialing

means that the number drawn from the directory has the last digit in the number replaced by a random number. This ensures that both listed and unlisted numbers are included in the sample

Independent samples

samples are independent if they are drawn randomly from different populations. For the purpose of analysis, data pertaining to different groups of respondents, e.g., males and females, are generally treated as independent samples.

Paired samples

samples are paired when the data for the two samples relate to the same group of respondents.

sample frame

some master list of all the members of the population

Sample frame error

the extent to which the sample frame does not perfectly match the population due to misrepresentation, overrepresentation, and/or underrepresentation. - It is the researchers responsibility to seek out a sample frame with the least amount of error at a reasonable costs. The researcher should also apprise the client of the degree of sample frame error involved.

Confidence level

the probability that a confidence interval will include the population parameter

Confidence interval

the range into which the true population parameter will fall, assuming a given level of confidence.

Sample Size When Estimating a Mean

• Sample Size Formula When Estimating a Mean n = s2 z2/ e2 Where: n= sample size s=variability estimated by one standard deviation z= Standard error associated with chosen level of confidence (1.96 or 2.58) e= acceptable error (accuracy)

proportionate sample size

would occur if we allocated sample size based upon each stratum's proportionate share of the total population.

Judgment Samples

• A judgment sample is somewhat different from a convenience sample in concept because a judgment sample requires a judgment or "educated guess" as to who should represent the population • Subjectivity enters in here, and perhaps the judgment includes more members of the population than a convenience sample, still certain members of the population will not have a probability of being selected into the sample

Define the Target Population

• A population is all cases that meet designated specifications for membership in the group • A census is defined as an accounting of everyone in the population. • Researchers must be very clear and precise in defining the target population - Example - Target population: Households in the city limits of Richardson, TX, with one or more children under the age of 18 living at home

Probability Sampling Methods

• A random sample is one in which every member of the population has an equal chance, or probability, of being selected into the sample. • Sample methods that embody random sampling are often termed probability sampling methods, because the chance of selection can be expressed as a probability. • The four probability sampling methods are: - Simple random sampling - Systematic sampling - Stratified sampling - Cluster sampling

Parametric Tests

• A t statistic assumes that the variable is normally distributed and the mean is known (or assumed to be known) and the population variance is estimated from the sample. • Assume that the random variable X is normally distributed, with mean μ and unknown population variance σ2 which is estimated by the sample variance s2. • Then, t = (X̅− μ)/ s X̅ • t is distributed with n − 1 degrees of freedom. • The t distribution is similar to the normal distribution in appearance. Both distributions are bell-shaped and symmetric. • As the number of degrees of freedom increases, the t distribution approaches the normal distribution.

Probability Sampling: Stratified Sampling

• A two-step process in which the population is partitioned into subpopulations, or strata. • The strata should be mutually exclusive and collectively exhaustive in that every population element should be assigned to one and only one stratum and no population elements should be omitted. • Next, elements are selected from each stratum by a random procedure, usually SRS. • A major objective of stratified sampling is to increase precision without increasing cost. • The stratification variables should also be closely related to the characteristic of interest. • Are stratified samples more accurate than random simple samples, given a sample size n? • In the next slide notice that the answers to the research questions differ between the strata.

Commonly Used Probes

• Any other reason? • Anything else? • Could you tell me more about your thinking on that? • Repeat the question • What do you mean? • Which would be closer to the way you feel? • Why do you feel that way? • Would you tell me what you have in mind?

ethics in sampling

• Appropriate definitions of the population, sampling frame, and sampling technique are essential if the research is to be conducted and the findings are to be used ethically. • Probability sampling techniques should be used whenever the results are to be projected to the population. • When conducting research with small populations, as in business-to-business marketing or employee research, researchers must be sensitive to preserving the respondents' anonymity

Sample size calculation

• As long as our variance estimate is accurate, we will correctly predetermine the amount of accuracy we will have in our survey results. • By being able to predetermine how accurate your results will be, you can confidently conduct surveys to estimate values of interest and be assured as to the accuracy of the sample findings.

Guidelines for Coding Unstructured Questions:

• Category codes should be mutually exclusive and collectively exhaustive. • Only a few (10% or less) of the responses should fall into the "other" category. • Category codes should be assigned for critical issues even if no one has mentioned them. • Data should be coded to retain as much detail as possible

Evaluation of Field Workers

• Cost and Time. Interviewers can be compared in terms of total cost (salary and expenses) per completed interview. • Response Rates. Monitor response rates on a timely basis so that corrective action can be taken if these rates are too low. • Quality of Interviewing. To evaluate interviewers on the quality of interviewing, the supervisor must directly observe the interviewing process. • Quality of Data. The completed questionnaires of each interviewer should be evaluated for the quality of data.

Data Prep and Social Media

• Data generated by a large networked panel could be made accessible to members enabling the discussions to dynamically organize and reorganize within the panel. • Social media respondents co-create, and thus 'respondents' become 'participants' in a shared enterprise, retaining the rights to set the agenda rather than simply responding to it. • Data collection involves the Web scraping process that first "crawls" the website to locate and identify the discussion topics, topic ID, topic starter, and topic start date. • It then uses topic ID to download posts and messages by the topic. • It is important that when storing the messages in a database, quotes in posts from others' text be removed to prevent double counting. • Text coding and categorization involves human inspection of a random sample of text messages to understand the type of acronyms, shorthand, and terminologies used and get a feel for the data. The human coder then develops a coding and categorization scheme and assists in the computer categorization of text until coding results are satisfactory. • In text mining and visualization, the coded text data are interpreted by matching positive or negative comments with overlapping terms that link the positive and negative comments to mentions of the product/brand.

Selection of Field Workers

• Develop job specifications for the project, taking into account the mode of data collection. • Decide what characteristics field workers should have. • Recruit appropriate individuals.

Data prep and ethics

• Discarding respondents after analyzing the data raises ethical concerns, particularly if this information is not fully disclosed in the written report. • The procedure used to identify unsatisfactory respondents and the number of respondents discarded should be clearly disclosed. • Although interpretations, conclusions, and recommendations necessarily involve the subjective judgment of the researcher, this judgment must be exercised honestly, free from any personal biases or agendas of the researcher or the client.

Variability: p times q

• Effect of "High v Low" estimates of variability in the population • If survey respondents have very little (low) variability, then most will select one category and few will select the other (90% vs 10%). - Note that "low" variability (90*10) gives you a lower number in the formula's numerator. Therefore, n will be lower! • If there is high variability, as when no two respondents agree then there is a 50% 50 % split, p times q becomes 50 X 50, or 2,500, which is the largest possible p x q number possible - Again, since p x q is in the numerator of our formula we will have a higher n if we estimate variance (p*q) to be high.

Research Questions

• Example for the mid priced, electric SUV: • What are the demographics of those who are most likely to buy? • How much are potential customers willing to pay? • What alternative brands do these customers consider?

Guidelines and Principles for the Written Report

• Form and Format • Headings indicate the topic of each section • Subheadings should divide that information into segments • Visuals • Visuals are tables, figures, charts, diagrams, graphs, and other graphic aids • A table systematically presents numerical data or words in columns and rows • A figure translates numbers into graphical displays so that findings can be comprehended visually • Style is the way one writes a report: • Write in third person • Avoid long paragraphs • Use jargon sparingly • Use strong verbs • Use active voice • Eliminate extra words • Avoid unnecessary changes in tense • In sentences, keep the subject and verb close together • Use faultless grammar • Edit carefully

Elements of a Marketing Research Report

• Front section • Title page • Letter of authorization • Letter/memo of transmittal • Table of contents • List of illustrations • Executive summary • Main section • Introduction • Research Objectives and Questions • Research Methodology • Findings • Limitations • Conclusions and Recommendations • End section • Appendices • End notes/References

Sampling & Social Media

• General social media content available in the public domain may not be representative or even appropriate in all cases. • Instead of targeting an entire site, select sections of sites that suit the brand's profile. Careful screening can result in a more targeted and representative sample. • Narrow your search results by designing search queries that mine social media content with consumer-, category-, or brandrelated terms. • Use text analysis that detects age, gender, geography, or other characteristics that distinguish different types of voices and then filter the results to more accurately reflect your target population.

General Qualifications of Field Workers

• Health. Field workers must have the stamina required to do the job. • Outgoing. The interviewers should be able to establish rapport with the respondents. • Communicative. Effective speaking and listening skills are a great asset. • Pleasant appearance. If the field worker's physical appearance is unpleasant or unusual, the data collected may be biased. • Educated. Interviewers must have good reading and writing skills. • Experienced. Experienced interviewers are likely to do a better job.

How to Select a Representative Sample

• Sample size formulas are only applicable when we have a representative sample. • How we draw a sample, the sample plan, determines whether the sample is representative. • There are two major types of sampling plans: probability and nonprobability sampling plans

Online International Marketing Research

• Identification and access to the relevant sampling elements varies widely across countries. • A reliable sampling frame might not be available. Government data in many developing countries might be unavailable or highly biased. • Identification of the decision maker and the relevant respondent might have to be done on a country-by-country basis. • Equivalence of samples is a key issue in marketing research studies extending beyond the home country. • Probability sampling techniques are uncommon in international marketing research. Rather, there is a reliance on quota sampling for both consumer and industrial surveys. • Sampling techniques and procedures vary in accuracy, reliability, and cost. • To achieve comparability in sample composition and representativeness, it might be desirable to use different sampling techniques in different countries.

Interpreting the Results of a One-Way ANOVA

• If the null hypothesis of equal category means is not rejected, then the independent variable does not have a significant effect on the dependent variable. • On the other hand, if the null hypothesis is rejected, then the effect of the independent variable is significant. • A comparison of the category mean values will indicate the nature of the effect of the independent variable.

List of Illustrations

• If the report contains tables and/or figures, include in the table of contents a list of illustrations along with the page numbers on which they appear. • Tables are words or numbers that are arranged in rows and columns. • Figures are: • Graphs • Charts • Maps • Pictures

analysis of covariance (ANCOVA)

• If the set of independent variables consists of both categorical and metric variables, the technique is called analysis of covariance (ANCOVA). In this case, the categorical independent variables are still referred to as factors, whereas the metric-independent variables are referred to as covariates.

n-way analysis of variance

• If two or more factors are involved, the analysis is termed

Frequency Distribution

• In a frequency distribution, one variable is considered at a time. • A frequency distribution for a variable produces a table of frequency counts, percentages, and cumulative percentages for all the values associated with that variable.

Cluster Sampling

• In cluster sampling: the population is divided into subgroups, called "clusters." • If each cluster is representative of the population, one or a few clusters can be selected and a census can be performed. This is a one-step area sample. • If the clusters are not similar, more clusters can be selected and samples taken from each. This is a two-step area sample. • This is desirable when geographic areas need to be surveyed because it can lower research costs.

Training of Field Workers

• In making the Initial Contact - Interviewers should be trained to make opening remarks that will convince potential respondents that their participation is important. • In asking Questions 1. Be thoroughly familiar with the questionnaire. 2. Ask the questions in the order in which they appear in the questionnaire. 3. Use the exact wording given in the questionnaire. 4. Read each question slowly. 5. Repeat questions that are not understood. 6. Ask every applicable question. 7. Follow instructions, skip patterns, probe carefully. • In Probing - Some commonly used probing techniques: • Repeating the question. • Repeating the respondent's reply. • Using a pause or silent probe. • Reassuring the respondent. • Clarification. • Using objective/neutral questions or comments. • In recording Answers - to unstructured questions: • Record responses during the interview. • Use the respondent's own words. • Do not summarize or paraphrase the respondent's answers. • Include everything that pertains to the question objectives. • Include all probes and comments. • Repeat the response as it is written down. • Terminating the Interview - The respondent should be left with a positive feeling about the interview.

Stratified Sampling

• In marketing research it is common to work with populations that contain subgroupings. • Stratified sampling is appropriate when we expect each subgroup to respond to research questions differently. • When we divide the population into these sub groupings we form different strata; each subgroup represents a stratum. The researcher should use some basis for dividing the population into strata that results in different responses to the key question(s) across strata.

Statistically Adjusting the Data Weighting

• In weighting, each case or respondent in the database is assigned a weight to reflect its importance relative to other cases or respondents. • Weighting is most widely used to make the sample data more representative of a target population on specific characteristics. • Yet another use of weighting is to adjust the sample so that greater importance is attached to respondents with certain characteristics.

Sample Size and Ethics

• It is a market researcher's ethical responsibility to try to educate a client on the wastefulness of excessively large samples. • Unethical researchers may recommend very large samples as a way to increase their profits, which may be set at a percentage of the total cost of the survey.

International Field Work

• Local field workers are preferable because they are familiar with the local language and culture and can create an appropriate climate for the interview, being sensitive to the concerns of the respondents. • Local field work agencies are unavailable in many countries; therefore, it might be necessary to recruit and train local field workers or import trained foreign workers. • In many countries, interviewers tend to help respondents with the answers and select households or sampling units based on personal considerations rather than the sampling plan. • Interviewer cheating can be more of a problem in many foreign countries. • For these reasons, validation of field work is critical.

Data prep and Mobile Marketing Research

• Marketing Data preparation and analysis in mobile marketing research (MMR) is similar to that in online (Internet) based surveys. • Firms like Pollfish and MFour have developed their own software to conduct basic data analysis. The results are made available to the client on a customized dashboard. The raw data are also made available to the client, such as in an Excel file format. The client can then conduct additional analyses to draw further insights into the specific components of the problem.

Why Use Samples?

• Marketing researchers, when collecting primary data, typically rely on a sample because: - taking a census of everyone in a market is time consuming - very costly, - and often leads to measurement errors. • Using a sample can generate results that generalize the entire population.

Statistically Adjusting the Data - Scale Transformation and Standardization

• Scale transformation involves a manipulation of scale values to ensure comparability with other scales or otherwise make the data suitable for analysis. • A more common transformation procedure is standardization. Standardized scores, Zi, may be obtained as: Zi = (Xi - X line) Sx

Determine the Sample Size

• Size of the population has no bearing on the size of the sample • Desired variation, precision, and confidence drive the sample size - Variation is outside the researcher's control; it's an artifact of the population - Precision and Confidence are inversely related • The more similar the population elements, the few people needed regardless of how large the population is Other Considerations - Costs/Available Research Budget § Larger sample sizes can cost more to recruit - Type of Analysis to be Conducted § Minimum requirements of statistical techniques must be met - Past Research § Historical evidence of sample sizes in similar studies can be a good guide

Mobile sampling

• Some marketing research firms maintain panels of mobile respondents. All the nonprobability and probability sampling techniques that we have discussed in this chapter can be implemented if a panel of mobile users is maintained. • An alternate method of recruiting respondents, especially in business-to-business (b2b) MMR, is with client lists. • Additional recruitment methods include links placed on websites, QR (Quick Response) codes on products, URLs included in letters, via links of pop-ups, through social media, or as part of a newsletter.

Chi-Square Analysis of a Cross-Tabulation Table

• The Cross-tabulation table only depicts the two variables simultaneously; it is NOT a statistical test. • Chi-square (X2) analysis is the examination of frequencies for two categorical variables in a cross-tabulation table to determine whether the variables have a significant relationship. • Chi-square (X2) analysis is a statistical test which will tell us if there is a "consistent, systematic" relationship...a "significant" association.

Findings

• The findings section is the most important and most detailed portion of the report. • This section should be organized around the research objectives for the study.

Identify the Sampling Frame

• The list of population elements from which a sample (n) will be drawn • Commonly used sampling frames include - Customer databases - Lists developed by data compilers

Main section of Report

• The main section is the bulk of the report. It contains: • An introduction to the report • A description of how your research was performed • A presentation of your findings • A statement of limitations • A list of conclusions and recommendations

Statistics Associated with Frequency Distribution

• The mean, or average value, is the most commonly used measure of central tendency. • The median of a sample is the middle value when the data are arranged in ascending or descending order. If the number of data points is even, the median is usually estimated as the midpoint between the two middle values - by adding the two middle values and dividing their sum by 2. The median is the 50th percentile. • The mode is the value that occurs most frequently. It represents the highest peak of the distribution. The mode is a good measure of location when the variable is inherently categorical or has otherwise been grouped into categories. • The range measures the spread of the data. It is simply the difference between the largest and smallest values in the sample. Range = X largest − X smallest • The interquartile range is the difference between the 75th and 25th percentile. • The variance is the mean squared deviation from the mean. The variance can never be negative. - When data points are clustered around the mean the variance is small. When data points are scattered the variance is large. • The standard deviation is the square root of the variance. • The coefficient of variation is the ratio of the standard deviation to the mean expressed as a percentage, and is a unitless measure of relative variability • Skewness refers to the tendency of the deviations from the mean being larger in one direction than in the other. It can be thought of as the tendency for one tail of the distribution to be heavier than the other. Kurtosis is a measure of the relative peak or flatness of the curve defined by the frequency distribution. The kurtosis of a normal distribution is zero. If the kurtosis is positive, then the distribution is more peaked than a normal distribution. A negative value means that the distribution is flatter than a normal distribution.

Quota Samples

• The quota sample establishes a specific quota for various types of individuals to be interviewed. The quotas are determined through application of the research objectives and are defined by key characteristics used to identify the population • Often, quota sampling is used as a means of ensuring that convenience samples will have the desired proportions of different respondent classes, thereby reducing the sample selection error but not eliminating it.

Cross-Validation

• The regression model is estimated using the entire data set. • The available data are split into two parts, the estimation sample and the validation sample. The estimation sample generally contains 50-90% of the total sample. • The regression model is estimated using the data from the estimation sample only. This model is compared to the model estimated on the entire sample to determine the agreement in terms of the signs and magnitudes of the partial regression coefficients. • The estimated model is applied to the data in the validation sample to predict the values of the dependent variable, Ŷi, for the observations in the validation sample. • The observed values, Yi, and the predicted values, Ŷi, in the validation sample are correlated to determine the simple r 2. This measure, r 2, is compared to R2 for the total sample and to R2 for the estimation sample to assess the degree of shrinkage.

The Importance of the Marketing Research Report

• The research report should be a factual message that transmits: • research results • vital recommendations • conclusions, and other important information to the client, who in turn bases his or her decisions on the contents of the report. • Marketing research users, as well as marketing research suppliers, agree that reporting the research results is one of the most important aspects of the marketing research process. The marketing research report is the culmination of a lot of work by a lot of people over a long period of time. • It may be the only part of the project seen by the client!

International Data Preparation

• The researcher should ensure that the data have been prepared in a comparable manner across countries or cultural units. This means that comparable procedures must be followed for checking questionnaires, editing, and coding. • Certain adjustments might be necessary to make the data comparable across countries. For example, the data might have to be adjusted to establish currency equivalents or metric equivalents. • Transformation of the data might be necessary to make meaningful comparisons and achieve consistent results.

Coding Questionnaires

• The respondent code and the record number appear on each record in the data. • The first record could contain additional codes: project code, interviewer code, date and time codes, and validation code.

Probability Sampling: Systematic Sampling

• The sample is chosen by selecting a random starting point and then picking every ith element in succession from the sampling frame. • The sampling interval, i, is determined by dividing the population size N by the sample size n and rounding to the nearest integer. • Example: there are 100,000 elements in the population and a sample of 1,000 is desired. In this case the sampling interval, i, is 100. A random number between 1 and 100 is selected. If, for example, this number is 23, the sample consists of elements 23, 123, 223, 323, 423, 523, and so on.

Statistics Associated with One-Way ANOVA

• The strength of the effects of X (independent variable or factor) on Y (dependent variable) is measured by eta2 (h 2). The value of h 2 varies between 0 and 1. • The null hypothesis that the category means are equal in the population is tested by an F statistic based on the ratio of mean square related to X and mean square related to error. • Mean square is the sum of squares divided by the appropriate degrees of freedom.

Desired Accuracy: e

• The term e is the amount of sample error (desired accuracy) that will be associated with the survey results. • Like z, e is determined by the judgment of the researcher and the client. • Most client's are satisfied with an error level of ±5%. • e is used to indicate how close your sample finding, in this case a percentage, will be to the true population percentage if it were repeated many times.

Estimating Variability (s) in Populations

• There are three ways to estimate variability (indicated by one standard deviation s, in the population): 1. First, do you have a previous study on the same population from which we can calculate s? 2. Second, do you have a pilot study to calculate s? ...and, third... 3. When the first two choices are not available we estimate the range of values that may be derived from the question and divide this range by 6. (+/-3 standard deviations account for 99% of the area under the normal curve, so 6 standard deviations are synonymous with the range). By dividing the range by 6 we can estimate 1 s!

Limitations

• This section is an honest accounting of major aspects of the research that constrain or temper the findings and conclusions. • No research is faultless, but all research projects strive to be as accurate as possible. • It should note major issues.

Hypothesis Testing Step Five

• Using standard normal tables, the probability of obtaining a z value of 1.88 can be calculated. • The shaded area between - ¥ and 1.88 is 0.9699. Therefore, the area to the right of z = 1.88 is 1.0000 − 0.9699 = 0.0301. • Note, in determining the critical value of the test statistic, the area to the right of the critical value is either α for a one-tail test and α/2 for a two-tail test.

Statistically Adjusting the Data - Variable Respecification

• Variable re-specification involves the transformation of data to create new variables or modify existing variables. • Dummy variables are used for re-specifying categorical variables. A dummy variable is a numerical variable used in regression analysis to represent subgroups of a sample. - Example - a person is given a value of 0 if in a control group or a value of 1 if in the treatment group. • The general rule is that to re-specify a categorical variable with K categories, K-1 dummy variables are needed.

Nonparametric Tests 2 Independent Samples

• When the difference in the location of two populations is to be compared based on observations from two independent samples, and the variable is measured on an ordinal scale, the Mann-Whitney U test can be used. • In the Mann-Whitney U test, the two samples are combined and the cases are ranked in order of increasing size. • The test statistic, U, is computed as the number of times a score from sample or group 1 precedes a score from group 2. • If the samples are from the same population, the distribution of scores from the two groups in the rank list should be random. An extreme value of U would indicate a nonrandom pattern, pointing to the inequality of the two groups. For samples of less than 30, the exact significance level for U is computed. For larger samples, U is transformed into a normally distributed z statistic. This z can be corrected for ties within ranks. • We examine again the difference in the Internet usage of males and females. This time the Mann-Whitney U test is used. The results are given in Table 15.17. • One could also use the cross-tabulation procedure to conduct a chi-square test. In this case, we will have a 2´2 table. One variable will be used to denote the sample, and will assume the value 1 for sample 1 and the value of 2 for sample 2. The other variable will be the binary variable of interest.

The Effects of Incidence Rate and Nonresponse on Sample Size

• Whenever you calculate the sample size, you are computing the number of respondents you should have complete your survey. • Invariably, surveys run into difficulties that require an upward adjustment in terms of the size of your sample you should begin with, or order from a sampling firm.

Cross-Tabulation Analysis

• With cross-tabulation, two categorical variables are arranged in a cross-tabulation table (a table in which data are compared using a row-and-column format). • The intersection of a row and column is called a crosstabulation cell. • A cross-tabulation analysis accounts for all of the relevant label-to-label relationships and it is the basis for the assessment of statistical significance of the relationships.

Recommendations

• are suggestions for how to proceed based on the conclusions. • Unlike conclusions, recommendations may require knowledge beyond the scope of the research findings themselves (conditions within the company and industry for example). • Therefore, researchers should exercise caution when making recommendations. • The researcher and the client should determine prior to the study whether a report is to contain recommendations and build a working relationship that fosters useful recommendations.

Conclusions

• are the deductions and inferences that have come about based on the research findings.

Nonparametric Tests

• are used when the independent variables are nonmetric. • Like parametric tests, nonparametric tests are available for testing variables from one sample, two independent samples, or two related samples.

End Section

• comprises of appendices, which contain additional information the reader may refer to for further reading that is not essential to reporting the data. • Blank copy of survey • Summary statistics (frequencies, percentages) • Cross tabs and other relevant statistical analysis • Maps • pictures

Title Page

• contains four major items of information: • Title of the document • The organization/person(s) for whom the report was prepared • The organization/person(s) who prepared the report • Date of submission

Letter/Memo of Transmittal

• describes the general nature of the research in a sentence or two and identifies the individual who is releasing the report. • a letter of transmittal is used to release or deliver the research report to an organization for which the researcher is not a regular employee. • a memo of transmittal is used to deliver the research report within the researcher's own organization.

Research Methodology

• describes, in as much detail as necessary, how the research was conducted, including a description of the data collection method, questionnaire design, sample plan, sample size, and analysis.

Research Objectives

• follow the introduction as a separate section and should state why the research was undertaken. • Example: to identify the best target market for a mid priced electric SUV.

Table of Contents

• helps the reader locate information in the research report. • Each heading should read exactly as it appears in the text and should identify the number of the page in which it appears.

Executive Summary

• is an overview of your report. • It serves as a summary for the busy executive or a preview for the in-depth reader. • It provides an overview of the most useful information, including the research objectives, questions, who was surveyed, sampling used, sample size, major findings, conclusions and recommendations. • Usually in one page

Letter of Authorization

• is the marketing research firm's certification to do the project. • It is particularly helpful in large organizations because it provides other users of the report with the name, title, and department of the individual(s) who authorized the project.

Introduction

• the main part of the report begins with the introduction section which orients the reader to research. • It should contain a statement of the background situation leading to the problem, the statement of the problem, and a summary description of how the research process was initiated. • It should contain a statement of the general purpose of the report and also the specific objectives and research questions for the research.