Social Research Methods Alan Bryman Social Research Methods Chapter 7: The nature of quantitative research
What is a concept? Concepts are: Concepts are useful for: Building blocks of theory Labels that we give to elements of the social world Categories for the organization of ideas and observations (Bulmer, 1984) Concepts are useful for: Providing an explanation of a certain aspect of the social world Standing for things we want to explain Giving a basis for measuring variation Page 151
Why measure? To delineate fine differences between people, organizations, or any other unit of analysis To provide a consistent device for gauging distinctions To produce precise estimates of the degree of the relationship between concepts Page 152
Indicators of concepts Produced by the operational definition of a concept Less directly quantifiable than measures Common sense understandings of the form a concept might take Multiple-indicator measures concept may have different dimensions Pages 152 and 153 4
Why use more than one indicator? Single indicators may incorrectly classify many individuals Single indicators may capture only a portion of the underlying concept or be too general Multiple indicators can make finer distinctions between individuals Multiple indicators can capture different dimensions of a concept Pages 153 and 154
What does reliability mean? Stability is the measure stable over time? e.g. test–retest method Internal reliability are the indicators consistent? e.g. split-half method Inter-observer consistency is the measure consistent between observers? Key concept 7.3 Page 157
What does validity mean? Does the indicator measure the concept? It does if it has: Face validity (right for the concept?) Concurrent validity (supported by a relevant criterion today?) Predictive validity (likely to be supported by a relevant criterion tomorrow?) Construct validity (are useful hypotheses produced?) Convergent validity (supported by results from other methods? Pages 158 and 159 7
Causality Explanation Direction of causal influence Confidence why things are the way they are Direction of causal influence relationship between dependent & independent variables Confidence in the researcher's causal inferences Page 163
Generalization Can findings be generalized beyond the confines of the particular context? Can findings be generalized from sample to population? How representative are samples? Pages 163 and 164
Replication Minimizing contamination from researcher biases or values Explicit description of procedures Control of conditions of study Ability to replicate in differing contexts Pages 164 to 166
Replication a study of tipping Brewster and Lynn (2014) explored customer racial discrimination in tipping Replicated previous study by Lynn which found black restaurant workers were tipped less Limitations of previous study – a measure of service quality New study – service quality not a mediating variable – race of the diner was not significant in tipping behaviour The findings were similar Page 165
The process of quantitative research Figure 7.1, Page 150 12
Criticisms of quantitative research Failure to distinguish between objects in the natural world and social phenomena Artificial and spurious sense of precision and accuracy Lack of ecological validity reliance on instruments and measurements Static view of social life Pages 166 and 167 13
Is it always like this? Quantitative research design is an ideal-typical approach Useful as a guide of good practice But discrepancy between ideal type and actual practice of social research Pragmatic concerns mean that researchers may not adhere rigidly to these principles Page 167 14
One reason for the discrepancy between the ideal and typical approaches Quantitative research is usually deductive (operational definition of concepts) But measurements can sometimes lead to inductive theorising And this means the factors give rise to the concepts, rather than making them operational Bryman (1988:28) calls this ‘reverse operationism’ Page 167 15
…and another reason Published accounts of quantitative research rarely report evidence of reliability and validity (Podsakoff & Dalton, 1987) Researchers are primarily interested in the substantive content and findings of their research Running tests of reliability and validity may seem an unappealing alternative! But researchers remain committed to the principles of good practice Page 167 16