The hazards of predicting divorce without crossvalidation

Research output: Contribution to journalArticle

Abstract

Divorce prediction studies (e.g., Gottman, Coan, Carrere, & Swanson, 1998) suggest that couples' eventual divorce can be very accurately predicted from a number of different variables. Recent attention to these studies has failed to consider the need to crossvalidate prediction equations and to consider the prevalence of divorce in the population. We analyze archival data to demonstrate that accuracy and predictive value drops precipitously during crossvalidation. We conclude that results of studies without crossvalidation analyses should be interpreted with extreme caution, no matter how impressive the initial results appear to be.

Original languageEnglish (US)
Pages (from-to)473-479
Number of pages7
JournalJournal of Marriage and Family
Volume63
Issue number2
DOIs
StatePublished - May 2001

Fingerprint

divorce
Hazard
Divorce
Cross-validation
Prediction

Keywords

  • Crossvalidation
  • Divorce
  • Overfitting
  • Prediction studies

ASJC Scopus subject areas

  • Social Sciences (miscellaneous)
  • Sociology and Political Science

Cite this

The hazards of predicting divorce without crossvalidation. / Heyman, Richard E.; Smith Slep, Amy M.

In: Journal of Marriage and Family, Vol. 63, No. 2, 05.2001, p. 473-479.

Research output: Contribution to journalArticle

@article{a1c7227acda44f7985da6de1600fca0c,
title = "The hazards of predicting divorce without crossvalidation",
abstract = "Divorce prediction studies (e.g., Gottman, Coan, Carrere, & Swanson, 1998) suggest that couples' eventual divorce can be very accurately predicted from a number of different variables. Recent attention to these studies has failed to consider the need to crossvalidate prediction equations and to consider the prevalence of divorce in the population. We analyze archival data to demonstrate that accuracy and predictive value drops precipitously during crossvalidation. We conclude that results of studies without crossvalidation analyses should be interpreted with extreme caution, no matter how impressive the initial results appear to be.",
keywords = "Crossvalidation, Divorce, Overfitting, Prediction studies",
author = "Heyman, {Richard E.} and {Smith Slep}, {Amy M.}",
year = "2001",
month = "5",
doi = "10.1111/j.1741-3737.2001.00473.x",
language = "English (US)",
volume = "63",
pages = "473--479",
journal = "Journal of Marriage and Family",
issn = "0022-2445",
publisher = "Wiley-Blackwell",
number = "2",

}

TY - JOUR

T1 - The hazards of predicting divorce without crossvalidation

AU - Heyman, Richard E.

AU - Smith Slep, Amy M.

PY - 2001/5

Y1 - 2001/5

N2 - Divorce prediction studies (e.g., Gottman, Coan, Carrere, & Swanson, 1998) suggest that couples' eventual divorce can be very accurately predicted from a number of different variables. Recent attention to these studies has failed to consider the need to crossvalidate prediction equations and to consider the prevalence of divorce in the population. We analyze archival data to demonstrate that accuracy and predictive value drops precipitously during crossvalidation. We conclude that results of studies without crossvalidation analyses should be interpreted with extreme caution, no matter how impressive the initial results appear to be.

AB - Divorce prediction studies (e.g., Gottman, Coan, Carrere, & Swanson, 1998) suggest that couples' eventual divorce can be very accurately predicted from a number of different variables. Recent attention to these studies has failed to consider the need to crossvalidate prediction equations and to consider the prevalence of divorce in the population. We analyze archival data to demonstrate that accuracy and predictive value drops precipitously during crossvalidation. We conclude that results of studies without crossvalidation analyses should be interpreted with extreme caution, no matter how impressive the initial results appear to be.

KW - Crossvalidation

KW - Divorce

KW - Overfitting

KW - Prediction studies

UR - http://www.scopus.com/inward/record.url?scp=18044370816&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=18044370816&partnerID=8YFLogxK

U2 - 10.1111/j.1741-3737.2001.00473.x

DO - 10.1111/j.1741-3737.2001.00473.x

M3 - Article

VL - 63

SP - 473

EP - 479

JO - Journal of Marriage and Family

JF - Journal of Marriage and Family

SN - 0022-2445

IS - 2

ER -