Table 4 Nonparametric fuzzy regression discontinuity (RD) estimates—gender heterogeneity

MSE Optimal

 Bandwidth

3.1

3.7

2.9

3.1

2.2

3.1

3

3

 Conventional Point Estimate

-1.9145

2.3318

-0.0176

0.2491

-0.7408

0.3909

-1.9779

8.8329

 Conventional 95% CI

[-4.4509, 0.6220]

[-2.8742, 7.5378]

[-0.6954, 0.6601]

[-0.5570, 1.0551]

[-1.7973, 0.3158]

[-0.5211, 1.3028]

[-7.3309, 3.3751]

[-5.1022, 22.768]

 Robust Point Estimate

-2.0148

2.1443

-0.0902

0.3111

-0.9604

0.5148

-1.6777

9.8741

 Robust 95% CI

[-4.9348, 0.9052]

[-4.0769, 8.3654]

[-0.851, 0.6706]

[-0.6532, 1.2754]

[-2.1403, 0.2195]

[-0.5432, 1.5728]

[-7.9675, 4.612]

[-6.9727, 26.7210]

 Effective obs L|R

494|486

424|422

535|500

414|412

356|346

382|391

485|467

341|359

CER Optimal

 Bandwidth

2.1

2.5

1.9

2.1

1.4

2.1

2

2

 Conventional Point Estimate

-2.5612

1.245

0.0408

0.8143

-0.6328

0.5464

-0.5967

9.0667

 Conventional 95% CI

[-5.7264, 0.6039]

[-3.8770, 6.3670]

[-0.6157, 0.6973]

[-0.9944, 2.6229]

[-1.5267, 0.2612]

[-0.7560, 1.8488]

[-6.0474, 4.8540]

[-8.5039, 26.6373]

 Robust Point Estimate

-2.586

1.211

0.0117

0.8529

-0.7135

0.6038

-0.4623

9.5349

 Robust 95% CI

[-5.9322, 0.7602]

[-4.3294, 6.7514]

[-0.6937, 0.7171]

[-1.0747, 2.7805]

[-1.6615, 0.2346]

[-0.7824, 1.9900]

[-6.2896, 5.3650]

[-9.5101, 28.5799]

 Effective obs L|R

346|304

280|290

373|333

276|285

253|239

251|268

333|300

232|250

Obs L|R

1557|1484

1176|1057

1820|1672

1302|1251

1554|1651

1166|1221

1587|1533

1216|1124