library(haven) ### package for reading .dta files## Warning: package 'haven' was built under R version 4.1.3data <- as.data.frame(read_dta("C:/Stuffs/IIT/3rd Semester/Advanced Econ Lab/UYdata.dta"),header=TRUE)
View(data)library(RCT) ### Github library by Isidoro Garcia-Urquieta for RCT ## Warning: package 'RCT' was built under R version 4.1.3summary_statistics(data = data)## # A tibble: 88 x 12
## variable mean n `0` `0.05` `0.1` `0.25` `0.5` `0.75` `0.9` `0.95`
## <chr> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl>
## 1 google_km 19.7 3003 1.64 6.89 8.27 12.7 19.2 25.4 31.6 34.5
## 2 tehsil_e 3.07 2949 1 1 1 1 4 5 5 5
## 3 treatment 0.973 2843 0 0 0 0 1 2 2 2
## 4 noncomplia~ 0.0533 3003 0 0 0 0 0 0 0 1
## 5 treatment0~ 0.614 3003 0 0 0 0 1 1 1 1
## 6 treatment_h 0.307 3003 0 0 0 0 0 1 1 1
## 7 treatment_~ 0.307 3003 0 0 0 0 0 1 1 1
## 8 hhscr_8_b 0.672 3003 0 0 0 0 1 1 1 1
## 9 hhgender_1~ 0.0613 3003 0 0 0 0 0 0 0 1
## 10 mrtlstatus~ 1.14 3003 1 1 1 1 1 1 1 3
## # ... with 78 more rows, and 1 more variable: `1` <dbl>library(expss) ### library for extracting labels from the .dta file## Warning: package 'expss' was built under R version 4.1.3## Loading required package: maditr## Warning: package 'maditr' was built under R version 4.1.3##
## To aggregate several columns with one summary: take(mtcars, mpg, hp, fun = mean, by = am)##
## Use 'expss_output_viewer()' to display tables in the RStudio Viewer.
## To return to the console output, use 'expss_output_default()'.##
## Attaching package: 'expss'## The following objects are masked from 'package:haven':
##
## is.labelled, read_spsslapply(data, var_lab)## $villagecode_str
## [1] "2011census"
##
## $google_km
## NULL
##
## $tehsil_e
## [1] "Tehsil name"
##
## $treatment
## [1] "Treatment Assignment - C/H/H+S"
##
## $noncompliance_village
## [1] "Villages added to C from H/H+S"
##
## $treatment012
## [1] "Dummy for whether assigned any treament"
##
## $treatment_h
## [1] "Dummy for whether assigned to treament - health"
##
## $treatment_hs
## [1] "Dummy for whether assigned to treament - health+subsidy"
##
## $hhscr_8_b
## [1] "Have LPG connection"
##
## $hhgender_1_b
## [1] "Female headed hh."
##
## $mrtlstatus_1_b
## [1] "Marital Status"
##
## $hhnum_b
## [1] "Household size"
##
## $hh_caste_b
## [1] "caste"
##
## $fire_1_b
## [1] "Use firewood for cooking"
##
## $fire_9_b
## [1] "If collect firewood"
##
## $dung_1_b
## [1] "Use dungcakes for cooking"
##
## $lpg_1_b
## [1] "Use LPG for cooking"
##
## $cookarea_1_b
## [1] "Separate cooking room YesNo"
##
## $cookarea_2_b
## [1] "Chimney/outlet YesNo"
##
## $age_pc_b
## [1] "Age of primary cook"
##
## $occu_b
## [1] "Hh. head self-employed or salaried"
##
## $obcpop_b
## [1] "OBC"
##
## $hindu_b
## [1] "Hindu"
##
## $ashacred_b
## [1] "Trust info. from ASHA"
##
## $education_b
## [1] "Pvt. primary school"
##
## $educationmid_b
## [1] "Govt. middle school"
##
## $healthstatus_b
## [1] "Primary health sub center"
##
## $irrigation_b
## [1] "Proportion of irrigated land"
##
## $road_b
## [1] "All weather road"
##
## $edu_hhhead_b
## [1] "Household head edu. above primary"
##
## $pc_edu_b
## [1] "Primary cook's edu. above primary"
##
## $scst_b
## [1] "SC/ST"
##
## $distance_dealer_close_b
## [1] "Closest distance between each hh to it nearest serving LPG dealer (based on geod"
##
## $fire_1_e
## [1] "Firewood"
##
## $fire_9_e
## [1] "Firewood collection YesNo"
##
## $dung_1_e
## [1] "Dung cakes"
##
## $cookarea_1_e
## [1] "Separate cooking room YesNo"
##
## $cookarea_2_e
## [1] "Chimney/outlet YesNo"
##
## $subdistricthqdist_c
## [1] "Sub District Head Quarter (Distance in km)"
##
## $bv85_c
## [1] "Lighting Electricity"
##
## $bv91_c
## [1] "Have latrine within house"
##
## $assets_index
## [1] "Household wealth index"
##
## $nonhindu
## NULL
##
## $totrefills_omc_e2
## NULL
##
## $totrefills_omc_b2
## [1] "Total no. of LPG refills (annual)"
##
## $pcdecisionindex_std_b
## [1] "Index for PC's decision making/bargaining power (using PCA)"
##
## $edu_trt
## NULL
##
## $edu_trth
## NULL
##
## $edu_trths
## NULL
##
## $asset_trt
## NULL
##
## $asset_trth
## NULL
##
## $asset_trths
## NULL
##
## $pcdecision_trt
## NULL
##
## $pcdecision_trth
## NULL
##
## $pcdecision_trths
## NULL
##
## $distdlr_imputed_trt
## NULL
##
## $distdlr_imputed_trth
## NULL
##
## $distdlr_imputed_trths
## NULL
##
## $distdlr_hybrid_b
## NULL
##
## $distdlr_hybrid_trt
## NULL
##
## $distdlr_hybrid_trth
## NULL
##
## $distdlr_hybrid_trths
## NULL
##
## $last_lpg_e
## NULL
##
## $last_lpg_b
## NULL
##
## $last_chulha_e
## NULL
##
## $last_chulha_b
## NULL
##
## $last_induction_e
## NULL
##
## $last_induction_b
## NULL
##
## $fueluse_index_b
## [1] "Index of Fuel Used in Last Meal"
##
## $fueluse_index_e
## [1] "Index of Fuel Used in Last Meal"
##
## $fire_visitb
## NULL
##
## $fire_visite
## NULL
##
## $dungmakeorcollect_b
## [1] "HH either makes dungcake or collects dung or does both"
##
## $dungmakeorcollect_e
## [1] "HH either makes dungcake or collects dung or does both"
##
## $dungmakeorcollectvisits_b
## [1] "HH makes visits for making dung or collecting dung or both"
##
## $dungmakeorcollectvisits_e
## [1] "HH makes visits for making dung or collecting dung or both"
##
## $induction_use_e
## [1] "Use induction stove for cooking"
##
## $induction_use_b
## [1] "Use induction stove for cooking"
##
## $monthly_winb
## [1] "No. of LPG refills per month (winter)"
##
## $monthly_sumb
## [1] "No. of LPG refills per month (summer)"
##
## $monthly_wetb
## [1] "No. of LPG refills per month (monsoon)"
##
## $att_missing
## NULL
##
## $treatment01
## NULL
##
## $treatment02
## NULL
##
## $treatment12
## NULL
##
## $hhhead_mrtlstatus
## [1] "Household head is married"
##
## $firewood_qty_b
## [1] "Qty. of firewood purchased last month (kg)"
##
## $dung_qty_b
## [1] "Qty. of dung cakes purchased last month"
##
## $`_merge`
## NULLdata <- data[!(data$att_missing==1),] ### dropping all the rows where attribution is absent##Table 1 is a timeline of the survey so I am skipping it
##Table 2: Balance of household characteristics at baseline.
df <- subset(data,select = c("hhnum_b", "hhgender_1_b", "age_pc_b", "edu_hhhead_b", "pc_edu_b", "hhhead_mrtlstatus", "occu_b", "scst_b", "obcpop_b", "hindu_b", "assets_index", "ashacred_b","treatment"))
balance_table(data = df,treatment = "treatment") ### Covariate balance table## # A tibble: 12 x 6
## variables1 Media_control1 Media_trat1 Media_trat2 p_value1 p_value2
## <chr> <dbl> <dbl> <dbl> <dbl> <dbl>
## 1 age_pc_b 34.2 33.9 33.6 0.611 0.267
## 2 ashacred_b 0.830 0.813 0.839 0.326 0.587
## 3 assets_index 1.55 1.63 1.51 0.0192 0.211
## 4 edu_hhhead_b 0.416 0.430 0.368 0.553 0.0314
## 5 hhgender_1_b 0.0611 0.0551 0.0732 0.579 0.297
## 6 hhhead_mrtlstatus 0.930 0.928 0.926 0.906 0.737
## 7 hhnum_b 6.13 6.15 6.17 0.853 0.738
## 8 hindu_b 0.930 0.931 0.890 0.945 0.00288
## 9 obcpop_b 0.435 0.421 0.436 0.549 0.970
## 10 occu_b 0.511 0.534 0.493 0.330 0.439
## 11 pc_edu_b 0.372 0.364 0.343 0.724 0.188
## 12 scst_b 0.393 0.407 0.428 0.542 0.125##Table 3: Balance of household fuel consumption at baseline.
df2 <- subset(data,select = c("fire_1_b","lpg_1_b","dung_1_b","induction_use_b","firewood_qty_b","dung_qty_b","hhscr_8_b","totrefills_omc_b2","monthly_winb","monthly_sumb","monthly_wetb","treatment"))
balance_table(data = df2,treatment = "treatment")## # A tibble: 11 x 6
## variables1 Media_control1 Media_trat1 Media_trat2 p_value1 p_value2
## <chr> <dbl> <dbl> <dbl> <dbl> <dbl>
## 1 dung_1_b 0.874 0.886 0.874 0.395 0.994
## 2 dung_qty_b 20.5 38.3 32.7 0.0709 0.00465
## 3 fire_1_b 0.755 0.728 0.763 0.183 0.679
## 4 firewood_qty_b 9.43 15.8 12.4 0.145 0.309
## 5 hhscr_8_b 0.643 0.705 0.672 0.00407 0.181
## 6 induction_use_b 0.0560 0.0761 0.0532 0.0803 0.790
## 7 lpg_1_b 0.719 0.771 0.738 0.00986 0.344
## 8 monthly_sumb 0.279 0.299 0.306 0.255 0.118
## 9 monthly_wetb 0.299 0.324 0.321 0.121 0.190
## 10 monthly_winb 0.273 0.285 0.276 0.435 0.845
## 11 totrefills_omc_b2 3.12 3.33 3.30 0.184 0.260library(sjPlot)## Warning: package 'sjPlot' was built under R version 4.1.3## Install package "strengejacke" from GitHub (`devtools::install_github("strengejacke/strengejacke")`) to load all sj-packages at once!\[ Y_{i v}^1=\beta_c+\beta_T T_v+\beta_0 Y_{i v}^0+\beta_X^{\prime} \mathbf{X}_{i v}+\beta_Z^{\prime} \mathbf{Z}_v+\varepsilon_{i v} \quad (Overall\ Treatment) \] \[ Y_{i v}^1=\beta_c+\beta_T^h T_v^h+\beta_T^{h s} T_v^{h s}+\beta_0 Y_{i v}^0+\beta_X^{\prime} \mathbf{X}_{i v}+\beta_Z^{\prime} \mathbf{Z}_v+\nu_{i v} \quad (Segregated \ effect) \]
##Table 4 (1) non-FE
reg1 <- lm(totrefills_omc_e2~ totrefills_omc_b2 +treatment012 , data =data)
summary(reg1)##
## Call:
## lm(formula = totrefills_omc_e2 ~ totrefills_omc_b2 + treatment012,
## data = data)
##
## Residuals:
## Min 1Q Median 3Q Max
## -13.1429 -1.0632 -0.7271 1.1277 15.3186
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 1.00597 0.08397 11.980 <2e-16 ***
## totrefills_omc_b2 0.80913 0.01290 62.721 <2e-16 ***
## treatment012 0.05721 0.09145 0.626 0.532
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 2.279 on 2726 degrees of freedom
## (62 observations deleted due to missingness)
## Multiple R-squared: 0.591, Adjusted R-squared: 0.5907
## F-statistic: 1970 on 2 and 2726 DF, p-value: < 2.2e-16##Table 4 (2) non-FE
reg2 <- lm(totrefills_omc_e2~totrefills_omc_b2 +treatment_h +treatment_hs,data = data)
summary(reg2)##
## Call:
## lm(formula = totrefills_omc_e2 ~ totrefills_omc_b2 + treatment_h +
## treatment_hs, data = data)
##
## Residuals:
## Min 1Q Median 3Q Max
## -13.1433 -1.1088 -0.7272 1.1392 15.3648
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 1.00585 0.08398 11.978 <2e-16 ***
## totrefills_omc_b2 0.80916 0.01290 62.721 <2e-16 ***
## treatment_h 0.01100 0.10641 0.103 0.918
## treatment_hs 0.10299 0.10615 0.970 0.332
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 2.279 on 2725 degrees of freedom
## (62 observations deleted due to missingness)
## Multiple R-squared: 0.5911, Adjusted R-squared: 0.5907
## F-statistic: 1313 on 3 and 2725 DF, p-value: < 2.2e-16##Table 4 (3) FE
reg3 <- lm(totrefills_omc_e2~ totrefills_omc_b2 +treatment012 +hhnum_b+ edu_hhhead_b+ occu_b +age_pc_b +pc_edu_b +hh_caste_b + nonhindu + assets_index +irrigation_b +education_b+healthstatus_b +road_b + subdistricthqdist_c, data =data)
tab_model(reg3,terms = c("totrefills_omc_b2","treatment012"))| totrefills_omc_e2 | |||
|---|---|---|---|
| Predictors | Estimates | CI | p |
| totrefills_omc_b2 | 0.81 | 0.78 – 0.83 | <0.001 |
|
Dummy for whether assigned any treament |
0.04 | -0.15 – 0.23 | 0.664 |
| Observations | 2498 | ||
| R2 / R2 adjusted | 0.590 / 0.588 | ||
##Table 4 (4) FE
reg4 <- lm(totrefills_omc_e2~ totrefills_omc_b2 +treatment_h +treatment_hs +hhnum_b+ edu_hhhead_b+ occu_b +age_pc_b +pc_edu_b +hh_caste_b + nonhindu + assets_index +irrigation_b +education_b+healthstatus_b +road_b + subdistricthqdist_c, data =data)
tab_model(reg4,terms = c("totrefills_omc_b2","treatment_h","treatment_hs"),dv.labels = "Impact of information campaign on annual LPG refill consumption.")| Impact of information campaign on annual LPG refill consumption. | |||
|---|---|---|---|
| Predictors | Estimates | CI | p |
| totrefills_omc_b2 | 0.81 | 0.78 – 0.83 | <0.001 |
|
Dummy for whether assigned to treament - health |
0.00 | -0.22 – 0.22 | 0.985 |
|
Dummy for whether assigned to treament - health+subsidy |
0.08 | -0.14 – 0.31 | 0.457 |
| Observations | 2498 | ||
| R2 / R2 adjusted | 0.590 / 0.588 | ||
##Table 5 Data is missing , contacted the author and verified