Relationships Among Ratio Variables with Common Components: Fact or Artifact
This paper examines relationships among ratio variables with common components. Simulation experiments are reported in which data samples with known relationships between ratios are analyzed. Observed ratio correlations are then compared to the known relationships. The simulation findings tentativel...
Saved in:
Published in | Social forces Vol. 58; no. 4; pp. 1109 - 1128 |
---|---|
Main Authors | , |
Format | Journal Article |
Language | English |
Published |
Chapel Hill, N.C
The University of North Carolina Press
01.06.1980
University of North Carolina Press |
Subjects | |
Online Access | Get full text |
Cover
Loading…
Summary: | This paper examines relationships among ratio variables with common components. Simulation experiments are reported in which data samples with known relationships between ratios are analyzed. Observed ratio correlations are then compared to the known relationships. The simulation findings tentatively suggest the absence of bias or spurious correlation. The sharp contrast between our findings and the results of simulation experiments carried out by ratio variable critics is examined. Finally, based on a more general mathematical argument, ratio correlations with common components are shown to be a valid form of analysis. |
---|---|
Bibliography: | istex:C7FF956C089206D5C0E68B42844F5CBB422EB0EB ark:/67375/HXZ-NQQZ5DR4-D ObjectType-Article-1 SourceType-Scholarly Journals-1 ObjectType-Feature-2 content type line 23 |
ISSN: | 0037-7732 1534-7605 |
DOI: | 10.1093/sf/58.4.1109 |