Abstract
With the quantitative–qualitative debate as a basis, this paper considers the relative merits and limitations of applying quantitative methods to comparative housing research. Although the initial discussion takes place in general terms, the example of the impact of an increase in real house prices on real consumers’ expenditure is employed to illustrate the issues that a housing researcher faces when applying statistical techniques to cross-country data. After reviewing the criticisms of the methodology, the paper concludes that quantitative methods should still be the workhorse for comparative housing research, but only if various conditions are satisfied and that the procedures are correctly followed.
Acknowledgements
The views expressed in this paper are a direct response of my exposure to qualitative techniques by colleagues on two EU-funded projects, OSIS and DEMHOW. They should not be held responsible for my limited understanding of the various qualitative approaches employed in these projects or for any errors in the logic of my arguments. I would also like to acknowledge the useful comments and suggestions from the two anonymous referees. The usual disclaimers apply.
Notes
1. See Horsewood and Neuteboom (2006) for an overview of the quantitative research and Elsinga et al. (2008) for a summary of the qualitative findings.
2. Experimental economics is a relatively new area for research and is frequently used to examine the theoretical foundations of microeconomics. Policy-based subjects, for instance environmental economics, can have a qualitative aspect to the research to obtain information on the preferences of the population.
3. The account of the exchange was cited in McAleer et al. (1985).
4. The quantitative-qualitative debate continues to rubble on, see Dawson et al. (Citation2006), Westerman and Yanchar (Citation2011).
5. See Teddlie and Tashakkori (2003) and Castro et al. (Citation2010) for an account of mixed methods research.
6. See Haffner et al. (Citation2010) for a more detailed discussion of universalistic assumptions in quantitative comparisons.
7. See Quilgars et al. (Citation2009) for an account of the procedures undertaken to select households to be interviewed
8. For further discussion of such issues, see principle component analysis in any econometric textbook.
9. For an account of the various time series approaches see Leamer (Citation1983), McAleer et al. (1985) and Pagan (1987).
10. IID(0, σ2) denotes that the errors are independent and identically distributed with a mean of zero and a constant variance σ 2.
11. For further information see Johansen (Citation1991) and Engle and Granger (Citation1987).
12. The research topic was covered in the European Journal of Housing Policy by Pacheco and Barata (Citation2005). See Kassanis (2007) for a correct application of techniques and an appropriate interpretation of the results.
13. See Kramer (1983).
14. In certain circumstances, taking logarithms can reduce the problem of heteroscedasticity.