1993
DOI: 10.1007/bf00991920
|View full text |Cite
|
Sign up to set email alerts
|

Statistical alternatives for studying college student retention: A comparative analysis of logit, probit, and linear regression

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
2
1
1
1

Citation Types

0
62
0

Year Published

1999
1999
2018
2018

Publication Types

Select...
4
2
2

Relationship

0
8

Authors

Journals

citations
Cited by 96 publications
(63 citation statements)
references
References 6 publications
0
62
0
Order By: Relevance
“…Although providing an improvement over the Chi Square test, Pierson (2002) estimates a model of volunteering -represented by a dichotomous dependent variableusing ordinary least squares (OLS) regression. It has been demonstrated (Dey & Astin, 1993) that logit, probit, and OLS regression analyses produce similar results when estimating a model with a dichotomous dependent variable, yet the assumptions of OLS regression are violated when the dependent variable is not continuous, and coefficients produced by OLS regression are often uninterpretable because they may produce predicted probabilities that extend beyond the logical boundaries of 0 and 1 (Cabrera, 1992). Although they presumably use logistic regression, Astin and Sax (1998) and Astin, et al (2000) do not report the statistical models that produced their findings.…”
Section: Literature Reviewmentioning
confidence: 99%
“…Although providing an improvement over the Chi Square test, Pierson (2002) estimates a model of volunteering -represented by a dichotomous dependent variableusing ordinary least squares (OLS) regression. It has been demonstrated (Dey & Astin, 1993) that logit, probit, and OLS regression analyses produce similar results when estimating a model with a dichotomous dependent variable, yet the assumptions of OLS regression are violated when the dependent variable is not continuous, and coefficients produced by OLS regression are often uninterpretable because they may produce predicted probabilities that extend beyond the logical boundaries of 0 and 1 (Cabrera, 1992). Although they presumably use logistic regression, Astin and Sax (1998) and Astin, et al (2000) do not report the statistical models that produced their findings.…”
Section: Literature Reviewmentioning
confidence: 99%
“…Logistic regression, probit analysis, simple linear regression and multiple regression are common statistical procedures when examining college student persistence. From a statistical standpoint, logistic, probit, and linear regression analyses are techniques that can be utilized to study and understand college student persistence (Dey & Astin, 1993). Logistic, probit, simple linear regression and multiple regression are associated with prediction.…”
Section: Discussionmentioning
confidence: 99%
“…Logistic regression was used to explain the effect that each of the control variables (i.e., sex, living arrangement, Pell Grant eligibility, and unweighted high school GPA) had on graduation. Because the outcome variable (1 = graduated, 0 = not graduated) was dichotomous, logistic regression was an appropriate technique for this analysis (Dey & Astin, 1993). and a higher percentage of off-campus students graduated in 6 years (6.56% compared to 5.93%).…”
Section: Results For Hypothesis Three (Graduation)mentioning
confidence: 99%
“…Because the outcome variable (1 = retained, 0 = not retained) is dichotomous, logistic regression was an appropriate technique for this analysis (Dey & Astin, 1993).…”
Section: Results For Hypothesis One (Retention)mentioning
confidence: 99%