How to construct an index for social research

How to create an index for social scientific research.

1. How to construct an index for social research? Mrunmayi Wadwekar PhD Scholar Maulana Azad National Institute of Technology, Bhopal

2. Index An index is a composite measure of variables, or a way of measuring a construct using more than one data item.  An index is an accumulation of scores from a variety of individual items. To create an index, we must select possible items, examine their empirical relationships, score the index, and validate it.

3. Construction of an Index

4. Item selection Selecting the items to be included in the index to measure the variable of interest. Validity - the item should measure what it is intended to measure. Unidimensionality - each item should represent only one dimension of the concept that is being measured. Deciding how general or specific the variable will be. Identifying the amount of variance that each item will provide. An index variable constitutes a scale measurement that is indicative of some hypothetical construct that can typically not be measured by a single question or item.

5. Examining Empirical Relationships An empirical relationship helps predict how the respondents will answer other questions based on answer of a particular question. If two items are empirically related to each other, we can argue that both items reflect the same concept and we can, therefore, include them in the same index.  To determine if the items are empirically related, cross-tabulations, correlation coefficients, or both may be used.

6. Index Scoring After finalizing the items to be included, scores/weights are assigned for particular responses, thereby making a composite variable out of the several items.  Unweighted aggregate index - each item score is weighted equally. Multivariate statistical techniques, such as exploratory factor analysis and principal component analysis could be considered in the construction of the index. Both methods work by assigning different weights to items through the calculation of factor scores. Weight assignment can be done by 4 ways: equal weights among items; theoretically categorized weights; schematic weights and variable weights.

7. Index Validation The index needs to be validated itself to make sure that it measures what it is intended to measure.  Construct validity is probably the most difficult to establish, as it is concerned with what the construct is ultimately measuring. Many variables, which are easily “observable” do not present any formidable difficulties in establishing construct validity. An index measure that is derived from less observable items, such as subjective evaluations or perceptions could be more challenging.

8. Problem of missing data Composite indices can deal with the problem of missing data in one of three ways. The first and simplest solution is case wise deletion. However, a very narrow indicator set or in absence of resources to collect primary data, case wise or indicator deletion is simply not feasible. The second technique is to impute missing values. Imputation is unreliable in cases where appropriate estimation models cannot be determined from available variables. Also, it can lead to highly erroneous results when data is missing for a very large number. A third approach is to use the existing data entirely and exclusively and to supplement this with an estimated margin of error, based on the number of missing items.

9. Water Quality Index Water Quality Index (WQI) is an universally accepted index to identify the quality of water and decide its use. WQI was proposed by R K Horton in 1965 and created by R M Brown (et al) in 1972. The calculation of the WQI is done using weighted arithmetic water quality index. where V i is the observed value of the i th parameter, S i is the standard permissible value of the i th parameter and V id is the ideal value of the i th parameter in pure water. All the ideal values (V id ) are taken as zero for drinking water except pH and dissolved oxygen.

10. Drinking water standards WQI Status 0-25 Excellent 26-50 Good 51-75 Poor 76-100 Very poor Above 100 Unsuitable for drinking Source: R M Brown (1972), C Chatterjee, M Raziuddin (2002)

11. Indian Standards: Drinking water 2005 The standard specifies the requirements, test methods and sampling procedures for ascertaining the suitability of water for drinking purpose. Drinking water should comply with requirements prescribed in various tables and also should comply with bacteriological, virological and biological requirements. Physical parameters – Colour, odour, Taste, Turbidity, Dissolved solids, pH, Total hardness Parameters concerning presence of metals such as copper, Iron, Manganese, Fluoride, Zinc, Aluminium etc. Presence of Toxic substances – Arsenic, Cyanide, Cadmium, Lead, Pesticides etc. and Radioactive substances. Presence of E. Coli, Total Coliform and thermotolerant coliform.

12. Air Quality index - SAFAR An air quality index is defined as an overall scheme that transforms the weighed values of individual air pollution related parameters (for example, pollutant concentrations) into a single number or set of numbers. Eight parameters (PM10, PM2.5, NO2 , SO2 , CO, O3 , NH3 , and Pb) having short-term standards have been considered for near real-time dissemination of AQI. It is recognized that air concentrations of Pb are not known in real-time and cannot contribute to AQI.

13. References Constructing a sophistication index as a method of market segmentation of commercial farming businesses in South Africa - Van Zyl, Hendrik Jacobus Dion, Gustav Puth , University of Pretoria Household Risk preparation indices – construction and diagnostics - Robert Foa , Harvard University Methodology of development of Social indices - Roberto Foa , Jeffery Tanner; World Bank Statistics for Management - Richard Levin, David Rubin, Sanjay Rastogi, Masood Husain Siddiqui