Abstract
This paper explored sample composition as a potential source of error in coefficient estimation and hypothesis testing by correlation with a dummy variable. Specific examples were provided to demonstrate to researchers the significance of this source of error. Although the problem is examined in light of the statistical consequences, composition of the sample, which is the actual source of error, is a methodological issue.
References
1.
Green
P. E.
Tull
D. J.
Research for marketing decisions . Englewood Cliffs, NJ : Prentice-Hall , 1978 .
2.
Johnston
J.
Economic methods . New York : McGraw-Hill , 1960 .
3.
Kerlinger
F. N.
Foundations of behavioral research . New York : Holt, Rinehart & Winston , 1973 .
4.
Kerlinger
F. N.
Pedhazur
E.
Multiple regression in behavioral research . New York : Holt, Rinehart & Winston , 1973 .
5.
Klainbaum
D. G.
Kupper
L. L.
Applied regression analysis and other multivariable methods . North Scituate, MA : Duxbury Press , 1978 .
6.
Murphy
J.
Introductory econometrics . Homewood, IL : Richard D. Irwin , 1973 .
7.
Nbter
J.
Wasserman
W.
Applied linear statistical models . Homewood, IL : Richard D. Irwin , 1974 .
8.
Nunnally
J. C.
Psychometric theory . New York : McGraw-Hill , 1967 .
9.
Orcutt
G. H.
Greenberger
M.
Microanalysis of socioeconomic systems: a simulation study . New York : Harper , 1961 .
10.
Roscoe
J. T.
Fundamental research statistics for the behavioral sciences . New York : Holt, Rinehart & Winston , 1974 .
11.
Suits
D.
The use of dummy variables in regression equations . Journal of the American Statistical Association , 1957 , 52 , 548 –551 .
12.
Thiel
H.
Economic forecasts and policy . Amsterdam : North-Holland , 1958 .