Table 4. Calorie consumption models.
| Model (1) | Model (2) | Model (3) | |
| Variables | Total calories (RE model) | Total calories (FE model) | Calories from more nutritious foods (FE model) |
| Bt area (ha) | 79.08*** (18.85) | 73.71*** (21.40) | 23.17** (10.05) |
| Farm size (ha) | 9.27** (4.22) | −0.69 (7.80) | 1.97 (3.56) |
| Education of farmer (years) | 9.41** (4.40) | – | – |
| Off-farm income (US$/year) | 0.07*** (0.02) | 0.05*** (0.02) | 0.01* (0.007) |
| Household size (AE) | −62.48*** (10.71) | −89.46*** (14.43) | −29.33*** (6.89) |
| Karnataka (dummy)a | 88.36 (57.97) | – | – |
| Andhra Pradesh (dummy)a | 21.46 (58.00) | – | – |
| Tamil Nadu (dummy)a | 212.86** (84.56) | – | – |
| 2004 (dummy)b | −34.35 (48.97) | −5.98 (51.60) | −45.25* (25.33) |
| 2006 (dummy)b | 13.68 (54.48) | 30.09 (61.12) | −112.87*** (29.41) |
| 2008 (dummy)b | −92.92 (60.51) | −74.59 (69.51) | −72.70** (30.20) |
| Constant | 3229.31*** (90.46) | 3537.08*** (78.16) | 843.23*** (41.42) |
| Number of observations | 1431 | 1431 | 1431 |
| R2 | 0.13 | 0.09 | 0.10 |
| Hausman test (chi-square statistic) | 16.82** | ||
The dependent variable in models (1) and (2) is the total number of kcal consumed per AE and day. The dependent variable in model (3) is the number of kcal consumed from more nutritious foods (i.e., pulses, fruits, vegetables, and animal products) per AE and day. All coefficient estimates can be interpreted as marginal effects; robust standard errors are shown in parentheses. AE: adult equivalent; RE: random effects; FE: fixed effects.
*
, **, ***Significant at the 10%, 5%, and 1% level, respectively.
a
The reference state is Maharashtra.
b
The reference year is 2002.