Heckman two-step estimates found no evidence of selection bias in the pretest–posttest gain score regressions based on camp attendance¶

Claim · All claims
Evidence · 1 study · causal r1 · q2
Currency · newest study 2016 · not yet checked against its sources · not reviewed by a person

Subclaims¶

q2 i? Mills ratio coefficients in all three Heckman specifications were statistically insignificant, indicating no detectable selection bias. → Bhattacharya et al. 2016

Evidence¶

Bhattacharya et al. 2016¶

Radha Bhattacharya, Andrew Gill, and Denise Stanley. (2016). The Effectiveness of Financial Literacy Instruction: The Role of Individual Development Accounts Participation and the Intensity of Instruction. Journal of Financial Counseling and Planning, Volume 27, Number 1, 2016, 20–35. https://doi.org/10.1891/1052-3073.27.1.20

q2 · i? · causal · r1

Heckman (1979) two-step estimator with a first-stage probit on camp attendance (N = 74 to 65) using present bias, young children, and car ownership as extra identifying variables; gain score models run on N = 50 to 43. No effect size is printed.

"The results in Table 5 show that we fail to find selection bias in the pretest to posttest regressions based on camp attendance, that is, the Mills ratio coefficients in all three specifications are statistically insignificant."

Discussion¶

Related Claims¶

• A 1-week financial fitness camp raised financial knowledge of low-income 8th graders by roughly 12 percentage points on a pretest–posttest design — related
• IDA participation and contribution level showed no statistically significant effect on financial knowledge gains from camp instruction — related
• Parents with high present bias were less likely to enroll their 8th graders in the financial literacy camp — related