The Use of Indexes in Professional Social Researches

by Luiza Mesesan-Schmitz (Author) Claudiu Coman (Author)

Summary

The book is written specifically for students or young researchers with no prior experience in the procedure of index construction. The book is designed as a self-teaching aid, thus some terms related to index are clarified, and some typologies of this kind of variable and the procedure of constructing index are presented. Using examples and datasets from social research, tables and SPSS screen captures, the readers are guided through this process. Some problems encountered in practice in this process of constructing an index are highlighted and some solutions are presented.

• Cover
• Title
• This eBook can be cited
• Contents
• Introduction
• On indexes or composite indices
• 1 Terminological observations
• 2 Composite measures
• 2.1 Defining indexes
• 2.2 Indexes versus scales
• 2.3 Indexes versus typologies
• 3 Procedure for index construction
• 3.1 Measurement of concepts that underpin indexes
• 3.2 Passing from indicators to indexes
• 3.2.1 Defining the concept
• 3.2.2 Establishing the range of possible values of the concept (valuing the index)
• 3.2.3 Creating an index using a statistical software (SPSS)
• a Statistical validation of the working operational definition
• b Statistical validation of the conceptual model for data aggregation and of the valuing of the index
• c The problem of missing data
• d Excluding missing data from the analysis5
• e Replacing missing data with other values
• f Index construction
• g Statistical validation of the index
• 3.3 Advantages and limits of using indexes
• 4 Types of indexes
• 4.1 Terminological observations
• 4.1.1 A typology of indexes
• a Summed score
• b Arithmetic mean score
• c Weighted arithmetic mean score
• d Ratio of scores index
• e Difference of scores index
• f Factor score
• g Indexes based on complex formulae
• 4.2 An illustration of a method for constructing indexes-summed or mean score12
• 4.2.1 Defining the concept
• 4.2.2 Establishing the range of possible values of the concept (valuing the index)
• 4.2.3 Statistical validation of the working operational definition
• 4.2.4 Statistical validation of the conceptual model for data aggregation and of the valuing of the index
• 4.2.5 Index construction
• 4.2.6 Statistical validation of the index
• 4.3 An illustration of a method for constructing indexes-count score
• 4.3.1 Defining the concept
• 4.3.2 Establishing the range of possible values of the concept (valuing the index)
• 4.3.3 Statistical validation of the working operational definition
• 4.3.4 Statistical validation of the conceptual model for data aggregation and of the valuing of the index
• 4.3.5 Index construction
• 4.3.6 Statistical validation of the index
• 4.4 An illustration of a method for constructing indexes-factorial score
• 4.4.1 Defining the concept
• 4.4.2 Establishing the range of possible values of the concept (valuing the index)
• 4.4.3 Statistical validation of the working operational definition
• 4.4.4 Statistical validation of the conceptual model for data aggregation and of the valuing of the index
• 4.4.5 Index construction
• 4.4.6 Statistical validation of the index
• The stability index in marketing studies
• 1 The concept of loyalty in marketing research
• 2 What does the index of stability16 represent in marketing research?
• 3 Advantages and limits of using the stability index
• 4 Empirical research where a stability index was used
• 4.1 The stability index construction – market research regarding beer consumption in Brasov and Prahova counties
• 4.2 Data analysis
• 4.2.1 Analysis of Ciucas beer brand consumer stability
• a The analysis of the data obtained with the stability index formed from three variables
• b Analysis of data obtained with the stability index from seven variables
• 4.2.2 The analysis of the stability of customers for competing brands
• Conclusion
• The stability index in election studies
• 1 What does the stability index represent in election studies?
• 2 Empirical research where a stability index was used
• 2.1 The stability index construction – a study regarding voting intentions in Brasov County
• 2.2 Data analysis
• 2.3 Optimising political communication with the help of a stability index
• Conclusion
• Appendixes
• Appendix 1 Orientation towards terminal values depending on the left-right scale (Correspondence analysis)
• Appendix 2 Orientation towards instrumental values depending on the left-right scale (Correspondence analysis)
• Appendix 3 Orientation towards terminal values (adjusted and standardised residuals)
• Appendix 4 Orientation towards instrumental values (adjusted and standardised residuals)
• Appendix 5 The perception upon problems in the locality (adjusted and standardised residuals) (adjusted and standardised residuals)
• Appendix 6 The main fears of the population (adjusted and standardised residuals)
• List of boxes
• List of figures
• List of tables
• References

←8 | 9→

Introduction

This work is based on ideas that we have discussed in other texts published before. The essential data support for the analyses included in this book is given by research that we coordinated over time in the political and marketing fields (1996–2008).

The gathering of these ideas into a book resulted from the desire to come to the support of young researchers who wish to initiate or perfect themselves in index construction. For this reason, the text has a didactic shade that brings to discussion several methodological aspects of constructing an index illustrated with concrete examples from our professional practice.

Thus, this work is structured on three chapters. The first chapter entitled: On indexes or composite indices, restates older ideas published in previous publications (Coman, 2011; Mesesan-Schmitz, 2011; Mesesan-Schmitz, 2015) where we discussed general terms about indexes, types of indexes and stages for constructing these variables. The text clarifies concepts related to indexes.

The second chapter, the one entitled: The stability index in marketing studies, restates an older theme we approached both in written materials (Coman & Tamasanu, 2006) and in our didactic or research activity (research contracts concluded with various institutions and organisations at a local level during 1996–2006). The material addresses aspects related to ways of constructing indexes illustrated with concrete examples used in marketing research. Essentially, the practical relevance of such an index is pointed out in the decisions related to relational marketing.

The last chapter entitled: The stability index in election studies, is based on ideas that we addressed in other published materials. (Coman, 2006; Coman, 2013)

This chapter illustrates briefly the calculation modality of a stability index in the field of election research and presents a series of particularities specific to this field. The stages to be passed are not presented in detail, these being mentioned in previous chapters, but enough methodological observations are made so that any reader could reconstruct such an index in a reliable manner.

In this work, we discuss a subject matter that is treated more peripherally in literature. When there is talk of the way of constructing an index, the discussion remains rather at a theoretical level regarding procedural stages that need to be followed and examples in this context are fairly schematic. The articles from Academic Journals have conditions related to the length of material and this is reflected in reduced information regarding technical and methodological ←9 | 10→aspects. As for books, the situation is not different. The opinion of Professor Babbie E. (2013, 201) is relevant in this respect. “An examination of actual social science research reports will show that researchers use indexes much more frequently than they do scales. Ironically, however, the methodological literature contains little if any discussion of index construction, whereas discussions of scale construction abound”.

In conclusion, this book represents an attempt to clarify a number of methodological problems related to constructing some indexes illustrated with examples from professional services that we conducted over time. Thus, we like to believe that we managed to pass beyond the theoretical aspects highlighted in methodological research works and we could point out problems encountered in practice in the process of constructing an index. A difficulty that young researchers meet, and not only them, consists in observing the theoretical conditions that they find in methodological books on one side and, on the other, coping with a series of practical limitations in regard to the context in which the research takes place and the demands of the customer who ordered the research.

Ideally, we could measure a concept by relating to a theory or by means of a standardised instrument such as scales. When one works as a sociologist in the academic environment, one has the freedom to design the research as one wishes within known parameters imposed only by previous knowledge relating to the respective issues. But what do we do when we find ourselves in the situation of sociologist practitioners and we have to conduct research in support of the professional service we provide and, besides theoretical conditioning, there are series of limits directly or indirectly imposed by the customer who ordered the research?

For instance, what do we do in the situation in which the customer imposes his own vision for measuring the concept, or when, for the concept we would like to measure, we have not identified a theory, or a standardised instrument, and temporal conditioning represents a factor that requires finding an ad-hoc solution?

Operationalisation is a solution at hand for each sociologist practitioner, but this procedure, in order to be valid, has to follow several rules (see chapter 1, subchapter 3.1), among which is the exhaustive character of the operational definition and, implicitly, a complete working operational definition (King, 2005, 202; Onut, 2014, 66–68). This definition will surely take a large part of the research instrument given that a question can measure a single indicator. But, as we sociologist practitioners know, besides this concept, our client wishes to measure, by means of the same instrument, many other concepts and we have to reach a compromise so that the instrument, namely the questionnaire, is not ←10 | 11→too long, but, at the same time, it ensures a measurement of concepts as precise and valid as possible.

If we wish to measure concepts with complex operational definitions, this will be reflected in a long questionnaire, and if the duration of a survey surpasses the optimum time, then superficial and erroneous answers can occur, and the number of random mistakes generated due to waning attention can increase. Unfortunately, as professors Traian Rotariu and Petru Ilut (2006, 140) mentioned, “this concept of limiting the duration of the enquiry, underlined with the insistence of all method manuals, seems to be ignored more and more often; huge questionnaires are encountered, especially in research initiated by various organisations.” Psychology manuals tell us that a person cannot be attentive to a matter for more than 15 minutes, while, from the experience that we have had as students and as field operators, the first signs of fatigue start to manifest in respondents after 30 minutes.

Therefore, we have to take decisions so that we do not generate significant errors due to the length of the questionnaire, while, at the same time, we have to take care not to distort the measurement of concepts too much and thus risking jeopardising the success of our professional services.

In this work, we try to illustrate two such situations, one in the field of marketing research and the other one in the field of political research and our approaches, in the respective situations, to solve the dilemmas mentioned above. Portions of the solutions mentioned are original and we have repeatedly tested and adapted them over time, in various research projects.

Details

Pages
164
Year
2020
ISBN (PDF)
9783631832202
ISBN (ePUB)
9783631832219
ISBN (MOBI)
9783631832226
ISBN (Softcover)
9783631827284
DOI
10.3726/b17422
Language
English
Publication date
2020 (August)
Keywords
Brand loyalty election campaigns factorial analysis cluster analysys statistical validity
Published
Berlin, Bern, Bruxelles, New York, Oxford, Warszawa, Wien, 2020. 164 pp., 19 fig. b/w, 57 tables.

166 pages