Enrolling at university and the social influence of peers

Nicolas Pistolesi

Accesso libero

Enrolling at university and the social influence of peers

Nicolas Pistolesi

| 01 feb 2022

IZA Journal of Labor Economics

Volume 11 (2022): Numero 1 (January 2022)

INFORMAZIONI SU QUESTO ARTICOLO

Articolo precedente

Articolo Successivo

Cita

Pubblicato online: 01 feb 2022

Pagine: -

Accettato: 02 nov 2021

DOI: https://doi.org/10.2478/izajole-2022-0001

Parole chiave
peer effects, higher education, university enrollment

© 2022 Nicolas Pistolesi, published by Sciendo

This work is licensed under the Creative Commons Attribution 4.0 International License.

Unemployment gap by education in France, 1987–2017.Notes: The figure represents the unemployment rate in France for individuals with and without postsecondary education. Source: Observatoire des inégalités https://www.inegalites.fr/Le-taux-de-chomageselon-le-diplome-et-l-age

University Enrollment and Share of Peers with C Feedback.Notes: The figure represents the relation between individual enrollment at university and the proportion of peers with C feedback. Dots are data averages over small intervals. The black continuous line represents the predicted value from a regression of the enrollment dummy variable on a polynomial function of order three of the share of peers with C feedback. The dashed grey lines represent 95% standard errors. The regression coefficients used to compute the predicted values are displayed in Table A3. The predicted value is computed from individual data and not data averages.

Balancing tests for the proportion of peers with C feedback.Notes: The figure represents the estimated coefficient from the separated regressions of each individual's observable characteristics on the share of peers with C feedback. The left panel displays OLS regressions, while the right panel presents fixed-effect regressions using major-by-high-school-by-year fixed effects and controlling by the peers mean characteristics; horizontal red lines are the standard errors.

Heterogeneous effects: Regressions by ability level

	Dep Var: Prob. to Enroll at University

	(1)	(2)	(3)	(4)	(5)
Prop. of peers with C feedback	−0.068^* (0.041)	−0.071^* (0.041)	−0.071^* (0.043)	−0.071^* (0.041)	−0.092^** (0.043)
Top 50% in ability distribution		0.010 (0.009)	0.010 (0.009)
Top 50% × Prop. of peers with C feedback			0.001 (0.121)
Bottom 25% in ability distribution				−0.081^*** (0.009)	−0.090^*** (0.011)
Top 25% in ability distribution				−0.079^*** (0.009)	−0.082^*** (0.009)
Bottom 25% × Prop. of peers with C feedback					0.023 (0.014)
Top 25% × Prop. of peers with C feedback					0.048^* (0.026)
Individual characteristics	✓	✓	✓	✓	✓
Peers characteristics	✓	✓	✓	✓	✓
Major-by-high-school-by-year fixed effects	✓	✓	✓	✓	✓
N	19,181	19,181	19,181	19,181	19,181
R²	0.023	0.023	0.023	0.030	0.030
Mean dep. var.	0.507	0.507	0.507	0.507	0.507
P-value F-test					0.033

Placebo regressions

Dep. Var: Probability to Enroll at University

	Peer mean computed in

	year t+1 (1)	year t+1 (2)	year t+2 (3)	year t+2 (4)
Prop. of peers with C feedback	−0.029 (0.048)	−0.048 (0.051)	−0.054 (0.049)	−0.056 (0.053)
Individual characteristics		✓		✓
Peers characteristics		✓		✓
Major-by-high-school-by-year fixed effects	✓	✓	✓	✓
	14,402	14,402	13,544	13,544
Mean dep. var.	0.508	0.508	0.505	0.505

Randomization check of high school class assignment

	(1)	(2)	(3)	(4)

	Percent Significant at the			Mean P-value

	10% level	5% level	1% level
Age	0.1	0.06	0.02	0.42
Male	0.12	0.06	0.02	0.47
Low social background	0.08	0.04	0.01	0.48
Optional courses in h-school	0.11	0.05	0.02	0.45
Held back in h-school	0.06	0.04	0.01	0.47
French Nationality	0.07	0.03	0.01	0.44

Balancing tests for the proportion of peers with C feedback

	(1)	(2)	(3)
Dependent variable
Age	0.013 (0.062)	−0.060 (0.069)	0.035 (0.079)
Male	0.352^*** (0.040)	0.157^*** (0.052)	0.087 (0.067)
Low soc. background	−0.068^*** (0.032)	−0.040 (0.033)	−0.034 (0.046)
Mean grade in h-school	−0.054^*** (0.020)	0.034 (0.025)	0.027 (0.028)
Optional course in h-school	−0.103^*** (0.050)	−0.060 (0.054)	−0.016 (0.050)
Held back in h-school	0.058^*** (0.026)	0.004 (0.028)	0.005 (0.034)
French nationality	0.002 (0.018)	0.041^* (0.025)	0.021 (0.033)
Major-by-high-school-by-year fixed effects		✓	✓
Peers characteristics			✓
Observations	19,181	19,181	19,181

Heterogeneous effects: Regressions by gender and by social background

	Panel A : Dep. Var: Prob. to Enroll

	Male		Female		All

	(1)	(2)	(3)	(4)	(5)

Prop. of peers with C feedback	−0.122^* (0.070)		−0.034 (0.063)
Prop. of male peers with C feedback		−0.110^** (0.054)		−0.013 (0.035)
Prop. of female peers with C feedback		−0.032 (0.035)		−0.035 (0.045)
Prop. of same gender peers with C feedback					−0.044^* (0.024)
Prop. of other gender peers with C feedback					−0.028 (0.024)
Individual Characteristics	✓	✓	✓	✓	✓
Peers Characteristics	✓	✓	✓	✓	✓
Major-by-high-school-by-year fixed effects	✓	✓	✓	✓	✓
Observations	7,891	7,891	11,290	11,290	19,181
R²	0.026	0.026	0.022	0.022	0.022
Mean dep. var.	0.509	0.509	0.506	0.506	0.507

	Panel B: Dep. Var.: Prob. to Enroll

	Low social background		High social background		All

	(1)	(2)	(3)	(4)	(5)

Prop of peers with C feedback	−0.105^* (0.062)		−0.027 (0.050)
Prop of low soc. back. peers with C feedback		−0.053^* (0.029)	0.021 (0.021)
Prop of high soc. back. peers with C feedback		−0.053 (0.044)	−0.019 (0.039)
Prop of same soc. back. peers with C feedback				−0.034 (0.022)
Prop of other soc. back. peers with C feedback				0.002 (0.018)
Individual characteristics	✓	✓	✓	✓	✓
Peers characteristics	✓	✓	✓	✓	✓
Major-by-high-school-by-year fixed effects	✓	✓	✓	✓	✓
Observations	5,068	5,068	14,113	14,113	19,181
R²	0.027	0.027	0.018	0.018	0.022
Mean dep. var.	0.535	0.535	0.497	0.497	0.507

Decomposition of the variance in peer group-level means

	Total peer mean Variance	Within h-school Majors (%)	Between h-school Majors (%)
Enrollment	0.09	46.48	53.52
C feedback	0.02	62.66	37.34
Age	0.76	8.82	91.18
Male	0.06	65.89	34.11
Low soc. background	0.03	60.21	39.79
Mean score in h-school	1.26	55.60	44.40
Elective class in h-school	0.09	47.06	52.94
Held-back in h-school	0.02	61.31	38.69
French nationality	0.04	17.86	82.14

OA policy and enrollment

Panel A: individual feedback

	A or B Feedback		C Feedback

Enrollment	0.51		0.47
	(0.01)		(0.01)

Panel B: peers’ feedback

Share of peers with C feedback	0.2	0.4	0.6	0.8
Enrollment	0.69	0.57	0.42	0.35
	(0.01)	(0.01)	(0.01)	(0.01)

Regression of individual enrollment on the share of peers with a C feedback

Dep. Var: Probability to Enroll at University

	(1)
Prop. of peers with C feedback	0.350 (0.104)
Prop.²	−2.513 (0.325)
Prop.³	1.892 (0.235)
Observations	19,181
R-squared	0.024

Regressions of individual enrollment on peers with C feedback from university application

	Dep. Var: Probability to enroll at university

	(1)	(2)	(3)	(4)	(5)	(6)
Prop. of peers with C feedback	−0.153^*** (0.048)	−0.094^** (0.037)	−0.095^** (0.037)	−0.055 (0.040)	−0.065 (0.041)	−0.068^* (0.041)
Individual characteristics		✓	✓		✓	✓
Peers characteristics			✓			✓
Major-by-high-school-by-year fixed effects				✓	✓	✓
Observations	19,181	19,181	19,181	19,181	19,181	19,181
R²	0.001	0.067	0.073	0.001	0.019	0.023
Mean dep. var.	0.507	0.507	0.507	0.507	0.507	0.507

Descriptive statistics

Panel A: Individual characteristics

	Mean	St. Dev.	Min.	Max.	Obs.

Age	18.04	1.03	16.69	23.18	19,181
Male	0.41	0.49			19,181
French nationality	0.92	0.25			19,181
Mean grade in h-school	11.81	2.06	3.25	19.33	19,181
Held back in h-school	0.07	0.26			19,181
Low social background	0.14	0.35			19,181
High achievers	0.32	0.46			19,181
Major in h-school: economics	0.44	0.49			19,181
Major in h-school: sciences	0.37	0.48			19,181
Major in h-school: literature	0.18	0.38			19,181
Public h-school	0.81	0.38			19,181
C feedback at univ. application	0.27	0.30			19,181
Enrolled at university	0.50	0.50			19,181

Panel B: High School Class Mean Characteristics

	Quantile

	Min	P25	P50	P75	Max

Mean class age	18.05	18.16	18.38	18.53	19.33
Class share of males	0.05	0.33	0.42	0.50	0.70
Class share of French	0.75	0.92	0.96	1.00	1.00
Class size	22	24	28	31	35
Mean class grade	10.10	11.25	11.79	12.29	14.17
Class share held back	0.00	0.00	0.05	0.09	0.38
Class share of low soc. back.	0.00	0.05	0.12	0.18	0.50
Class share high achievers	0.00	0.09	0.26	0.43	0.92
Class share enrolled at university	0.08	0.42	0.57	0.87	1.00
Class share observed	0.62	0.77	0.88	1.00	1.00

Heterogeneous effects: Regressions by number of observed peers

	Dep Var: Prob. to Enroll at University

	(1)	(2)	(3)	(4)
Prop. of peers with C feedback	−0.068^* (0.041)	−0.080 (0.051)	−0.061 (0.042)	−0.075^* (0.043)
Nb of obs. peers above the median		−0.002 (0.013)
Prop. of obs. peers with C feedback × Above median		0.044 (0.081)
Total class observed			−0.021 (0.022)
Prop. of obs. peers with C feedback × Total Class			−0.057 (0.188)
Prop. obs. peers >80%				−0.014 (0.016)
Prop. of obs. peers with C feedback × Observed peers >80%				0.087 (0.051)
Individual characteristics	✓	✓	✓	✓
Peers characteristics	✓	✓	✓	✓
Major-by-high-school fixed effects	✓	✓	✓	✓
N	19,181	19,181	19,181	19,181
R²	0.023	0.023	0.023	0.023
Mean dep. var.	0.507	0.507	0.507	0.507

eISSN:: 2193-8997
Lingua:: Inglese

Frequenza di pubblicazione:: Volume Open
Argomenti della rivista:: Business and Economics, Political Economics, Microeconomics, Macroecomics, Economic Policy, Mathematics and Statistics for Economists, Econometrics

Feed RSS della rivista

Enrolling at university and the social influence of peers

Pubblicato online: 01 feb 2022

Pagine: -

Accettato: 02 nov 2021

DOI: https://doi.org/10.2478/izajole-2022-0001

Parole chiave
peer effects, higher education, university enrollment

© 2022 Nicolas Pistolesi, published by Sciendo

This work is licensed under the Creative Commons Attribution 4.0 International License.

Figure 1

Figure 2

Figure 3

Heterogeneous effects: Regressions by ability level

Placebo regressions

Randomization check of high school class assignment

Balancing tests for the proportion of peers with C feedback

Heterogeneous effects: Regressions by gender and by social background

Decomposition of the variance in peer group-level means

OA policy and enrollment

Regression of individual enrollment on the share of peers with a C feedback

Regressions of individual enrollment on peers with C feedback from university application

Descriptive statistics

Heterogeneous effects: Regressions by number of observed peers

Enrolling at university and the social influence of peers

Pubblicato online: 01 feb 2022

Pagine: -

Accettato: 02 nov 2021

DOI: https://doi.org/10.2478/izajole-2022-0001

Parole chiavepeer effects, higher education, university enrollment

© 2022 Nicolas Pistolesi, published by Sciendo

This work is licensed under the Creative Commons Attribution 4.0 International License.

Figure 1

Figure 2

Figure 3

Heterogeneous effects: Regressions by ability level

Placebo regressions

Randomization check of high school class assignment

Balancing tests for the proportion of peers with C feedback

Heterogeneous effects: Regressions by gender and by social background

Decomposition of the variance in peer group-level means

OA policy and enrollment

Regression of individual enrollment on the share of peers with a C feedback

Regressions of individual enrollment on peers with C feedback from university application

Descriptive statistics

Heterogeneous effects: Regressions by number of observed peers

Parole chiave
peer effects, higher education, university enrollment