Current facets (Pre-Master)
One of the ways to deal with this problem is to take stock of available research in a particular field periodically. Techniques for doing this have considerably improved in the last decade; among other things because the growing piles of research findings have made this approach more profitable. Classic books on research synthesis include Wolf (1993), Light & Pillemer (1984) and Cooper and Hedges (1994), and since 2009 the scientific journal Research Synthesis Methods is available.
Research synthesis should not be mixed up with ‘secondary analysis’ of available data, that is, re-analysis of data initially gathered by someone else. Data-analysis, whether primary or secondary, produces ‘findings’, such as distributions of a particular variable in a particular population and correlations of such a variable with other variables. Research synthesis is about assembling such ‘findings’ from different studies to get a broader view and to reduce bias.
Traditionally research synthesis is done in periodical literature reviews, such as ‘state-of-the-art’ articles in scientific journals or in handbooks on a particular subject. Such studies are typically made by senior specialists and have sections on the development of the research field, summary of the main findings and an agenda for future research. This is called narrative research synthesis. The term ‘narrative’ denotes both that reviewers interpret the research findings for constructing a bigger story and also that they draw heavily on the interpretations found in the literature under review. Research synthesis of this kind is often theory driven and screens the available data for evidence for or against a particular theory.
More recently, techniques for quantitative meta-analysis have appeared on the scene (e.g. Hunter& Schmidt 2004). In this approach the focus is on observed facts rather than on interpretations. Quantitative meta-analysis aims at an accurate description of reality in the first place, both of general trends and of contextual variations. Today this approach is facilitated by new statistical tools and specialized computer programs (e.g. NCSS software for meta-analysis). This approach requires the existence of a large and homogenous body of research findings; as condition that is more often met in the medical field than in the social sciences.
Today some 1000 studies on happiness appear every year and the heap of findings has grown too big to be handled by narrative research reviews, but ever more suitable for quantitative meta-analysis. Yet meta-analysis requires investment in the gathering of relevant research and in homogenizing the findings. This investment is particularly high if one wants to cover the entire world's research. Such investments are made in capital-intensive fields such as pharmacological research, but uncommonly in the social sciences. The few meta-analyses of empirical happiness research have been based on small collections, e.g. Stock et. al. (1983). As yet, all the meta-analyses have been one-time shots, leaving no common database to build on. Each new investigator has to make a new start. Not surprisingly few do so.
World Database of Happiness
EHERO facilitates research synthesis on happiness by maintaining the World Database of Happiness, which is an ‘archive of research findings on subjective enjoyment of life’ Veenhoven 2011). The database restricts to findings on happiness in the sense of ‘life satisfaction’. Findings are described on an electronic ‘finding page’ using a standard format and terminology. To date the database includes some 25.000 such pages, which can be easily browed and combined in ‘reports’ on a particular topic.
State-of-the-art reviews on particular topics
Using the World Database of Happiness EHERO can produce reviews of the available knowledge on happiness in particular publics (such as the elderly) or on particular subjects (such as the relation with income).
Support the gathering of research findings on happiness in your field
Focussed reviews of happiness research can be made in commission or can be produced in the context of a longer sponsoring relationship.
- Cooper, H. & Hedges, L.V. (Eds.) (1994) Handbook of Research Synthesis Russell Sage Foundation, New York
- Hunter, J.E. & Schmidt, F.L. (2004) Methods of meta-analysis: Correcting error and bias in research findings Sage, New York
- Light, R.J. & David B. Pillemer, D. B. (1984) Summing up: the science of reviewing research, Harvard University Press
- NCSS Software for meta-analysis Available at: http://www.ncss.com/metaanal.html
- Veenhoven, R. (2011) The World Database of happiness: Example of a focussed findings archive, Working paper No. 169, German Data Forum RatSWD
- Wolf, F.M. (1993) Meta-analysis: Quantitative methods for research synthesis Sage, New York, USA