The process of selecting a data analysis strategy is described. The selection of a data analysis strategy should be based on the earlier steps of the marketing research process. known characteristics of the data. properties of statistical techniques, and the background and philosophy of the researcher.
Data analysis is not an end in itself. Its purpose is to produce information that will help address the problem at hand. The selection of a data analysis strategy must begin with a consideration of the earlier steps in the process: problem definition (Step I), development of an approach (Step II), and research design (Step III). The preliminary plan of data analysis prepared as part of the research design should ~>eused as a springboard, Changes may be necessary in light of additional information generate. n subsequent stages of the research process.
The next step is to consider the known characteristics of the data. The measurement scales used exert a strong influence on the choice of statistical techniques .ln addition. the research design ma) favor certain techniques. For example, analysis of variance is suited for analyzing experimental data from causal designs. The insights into the data obtained during data preparation can be valuable for selecting a strategy for analysis.
Earlier Steps (I, II, and III) of the Marketing Research Process
Known Characteristics of the Data
Properties of Statistical Techniques
Background and Philosophy of the Research
Data Analysis Strategy
It is also important to take into account the properties of the statistical techniques, particularly their purpose and underlying assumptions. Some statistical techniques are appropriate for examining differences in variables, others for assessing the magnitudes of the relationships between variables, and others for making predictions. The techniques also involve different assumptions. and some techniques can withstand violations of the underlying assumptions better than others. A classification of statistical techniques is presented in the next section.
Finally. the researcher’s background and philosophy affect the choice of a data analysis strategy. The experienced. statistically trained researcher will employ a range of techniques, including advanced statistical methods. Researchers differ in their willingness to make assumptions about the variables and their underlying populations. Researchers who are conservative about making assumptions will limit their choice of techniques to distribution-free methods. In general, several techniques may be appropriate for analyzing the data from a given project.
A Classification of Statistical Techniques
Statistical techniques can be classified as univariate or multivariate. Univariate techniques are appropriate when there is a single measurement of each clement in the sample, or there are several measurements of each element but each variable is analyzed in isolation. Multivariate techniques, on the other hand, are suitable for analyzing data when there are two or more measurements of each element and the variables are analyzed simultaneously. Multivariate techniques are concerned with the simultaneous relationships among two or more phenomena. Multivariate techniques differ from univariate techniques in that they shift the focus away from the levels (averages) and distributions (variances) of the phenomena, concentrating instead upon the degree of relationships (correlations or covariances) among these phenomena.I! The univariate and multivariate techniques are described in detail in subsequent chapters; here we show how the various techniques relate to each other in an overall scheme of classification.
Univariate techniques can be classified based on whether the data are metric or nonmetric. Metric data are measured on an interval or ratio scale. Nonmetric data are measured on a nominal or ordinal scale . These techniques can be further classified based on whether one, two. or more samples are involved. It should be noted that here the number of samples is determined based on how the data are treated for the purpose of analysis, not based on how the data were collected. For example, the data for males and females may well have been collected as a single sample, but if the analysis involves an examination of sex differences, two sample techniques will be used, The samples are independent if they are drawn randomly from different populations. For the purpose of analysis, data pertaining to different groups of respondents, for example, males and females. are generally treated as independent samples. On the other hand, the samples are paired when the data for the two samples relate to the same group of respondents.
For metric data, when there is only one sample, the z test and the I test can be used. When there are two or more independent samples, the z test and I test can be used for two samples, and one-way analysis of variance (one-way ANOVA) for more than two samples. In the case of two related samples, the paired I test can be used. For nonmetric data involving a single sample, frequency distribution, chi-square, Kolmogorov-Smirnov, runs, and binomial tests can be used. For two independent samples with nonmetric data, the chi-square, Mann- Whitney, Median, K-S, and Kruskal-Wallis one-way analysis of variance (K-W ANOVA) can be used. In contrast, when there are two or more related samples, the sign, Wilcoxon, McNemar, and chi-square tests should be used (see Figure 14.6)
Multivariate statistical techniques can be classified as dependence techniques or interdependence techniques (see Figure 14.7). Dependence techniques are appropriate when one or more variables can be identified as dependent variables and the remaining as independent variables. When there is only one dependent variable, cross-tabulation, analysis of variance and covariance. regression, two-group discriminant analysis, logit analysis. and conjoint analysis can be used. However. if there is more than one dependent variable, the appropriate techniques are multivariate analysis of variance and covariance, canonical correlation. multiple discnrninant analysis, logit analysis, structural equation modeling, and path analysis. In interdependence techniques. the variables are not classified as dependent or independent; rather, the whole set of interdependent relationships is examined. These techniques focus on either variable interdependence or inter-object similarity. The major techniques for examining variable interdependence are factor analysis and confirmatory factor analysis. Analysis of intersubject similarity can be conducted by cluster analysis and multidimensional scaling.