Conventional least-squares regression can lead to misleading results if the data have a hierarchical structure. An appropriate linear model is presented for such data. The model has conventional regression, variance-component, and random coefficient models as special cases and may be calibrated by use of recently available software. The effectiveness of the model is demonstrated by an analysis of earnings in the engineering industry. Particular attention is given to the problems of interpreting the parameter estimates and residuals.
Get full access to this article
View all access options for this article.
References
1.
AitkinMAndersonD AFrancisB JHindeJ P, 1988Statistical Modelling in GLIM3 (Oxford University Press, Oxford) forthcoming
2.
AitkinMLongfordN, 1986, “Statistical modelling issues in school effectiveness studies”Journal of the Royal Statistical Society, Series A1491–43
3.
BainJ O, (Ed.), 1983Industrial Relations in Britain (Oxford University Press, Oxford)
BlalockH M, 1960Social Statistics (McGraw-Hill, New York)
6.
ChamberlainG, 1984, “Panel data”, in Handbook of Econometrics, Volume 2 Eds GrilichesZIntriligatorM D, (North-Holland, Amsterdam) pp 1247–1318
7.
CookR DWeisbergS, 1982Residuals and Influence in Regression (Chapman and Hall, London)
8.
DempsterA PRubinD BIsutakawaR K, 1981, “Estimation in variance components models”Journal of the American Statistical Association76341–353
9.
FryerRMartinR, 1973Redundancy and Paternalist Capitalism (Allen and Unwin, Hemel Hempstead, Herts)
10.
GoldsteinH, 1986, “Multilevel mixed linear model analysis using iterative generalised least squares”Biometrika7343–56
11.
HartR AMcKayD I, 1975, “Engineering earnings in Britain, 1914–68”Journal of the Royal Statistical Society, Series A13832–50
12.
HildrethCHouckJ P, 1968, “Some estimates for a linear model with random coefficients”Journal of the American Statistical Association63584–595
13.
JeffreysH, 1983Theory of Probability3rd edition (Clarendon Press, Oxford)
14.
JohnsonL WHensherD A, 1979, “A random coefficient model of the determinants of frequency of shopping trips”Australian Economic Papers18322–336
15.
LindleyD VSmithA F M, 1972, “Bayes estimates for the linear model”Journal of the Royal Statistical Society, Series B341–41
16.
LongfordN T, 1985a, “Statistical modelling of data from hierarchical structures using variance component analysis”, in Generalized Linear Models Eds GilchristRFrancisBWhittakerJ, (Springer, Berlin) pp 112–119
17.
LongfordN T, 1985b, “Mixed linear models and an application to school effectiveness”Computational Statistics Quarterly2109–117
18.
LongfordN T, 1986VARCL Manual (Centre for Applied Statistics, University of Lancaster, Lancaster LA1 4YL, England, and Educational Testing Service, Rosedale Road-22T, Princeton, NJ 08541 USA)
19.
NorcliffeG B, 1977Inferential Statistics for Geographers (Hutchinson, London)
20.
PennR, 1983, “The course of wage differentials between skilled and nonskilled workers in Britain between 1856 and 1864”British Journal of Industrial Relations2169–90
21.
PennR, 1985Skilled Workers in the Class Structure (Cambridge University Press, Cambridge)
22.
PennR, 1986, “Trends in earnings of engineering workers in Rochdale, 1979–1984”, Lancaster-Manchester Universities Working Papers on Social Change and Economic Life, Number 2; copy from Department of Sociology, University of Lancaster, Lancaster LA1 4YL, England
23.
PennRMartinA MDaviesR B, 1987, “Changes in the differentials of engineering workers since 1979: An analysis of earnings data from Rochdale”British Journal of Industrial Relations forthcoming
24.
BrownH Phelps, 1977The Inequality of Pay (Oxford University Press, Oxford)
25.
PicklesA R, 1985An Introduction to Likelihood Analysis (Geo Books, Norwich)
26.
PicklesA RDaviesR B, 1988, “Inference from cross-sectional and longitudinal data for dynamic behavioural processes”, in Contemporary Developments in Quantitative Geography Eds HauerJTimmermansHWrigleyN, (D Reidel, Dordrecht) forthcoming
27.
RouthG, 1965Occupation and Pay in Great Britain, 1906–1960 (Cambridge University Press, Cambridge)
