1 / 10

Dummy Variables and Interactions

Dummy Variables and Interactions. Dummy Variables. What is the the relationship between the % of non-Swiss residents (IV) and discretionary social spending (DV) in Swiss municipalities? . reg def_social_head log_pctforeign if year==2005

noel
Download Presentation

Dummy Variables and Interactions

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. Dummy Variables and Interactions

  2. Dummy Variables What is the the relationship between the % of non-Swiss residents (IV) and discretionary social spending (DV) in Swiss municipalities? . regdef_social_headlog_pctforeign if year==2005 Source | SS df MS Number of obs = 1151 -------------+------------------------------ F( 1, 1149) = 0.54 Model | 155287.676 1 155287.676 Prob > F = 0.4631 Residual | 331138316 1149 288196.968 R-squared = 0.0005 -------------+------------------------------ Adj R-squared = -0.0004 Total | 331293604 1150 288081.395 Root MSE = 536.84 -------------------------------------------------------------------------------- def_social_h~d | Coef. Std. Err. t P>|t| [95% Conf. Interval] ---------------+---------------------------------------------------------------- log_pctforeign | 7.449417 10.14842 0.73 0.463 -12.4621 27.36093 _cons | 663.3708 27.61287 24.02 0.000 609.1935 717.5481 Data drawn from 6 different cantons (states).

  3. Dummy Variables . tabstatdef_social_headpctforeign, by(canton) Summary statistics: mean by categories of: canton canton | def_so~dpctfor~n ---------+-------------------- 1 | 504.8596 .135417 2 | 202.6653 .2730006 3 | 474.0729 .0919272 5 | 945.3242.0633038 6 | 377.7097 .1101544 7 | 110.8237.0918817 ---------+-------------------- Total | 501.6756 1.689802 ------------------------------

  4. Dummy Variables We can control for the fact that municipalities are drawn from different cantons by allowing the default expectation (intercept) for each canton to vary: . regdef_social_headlog_pctforeigni.canton if year==2005 Source | SS df MS Number of obs = 1151 -------------+------------------------------ F( 6, 1144) = 127.25 Model | 132602310 6 22100384.9 Prob > F = 0.0000 Residual | 198691294 1144 173681.201 R-squared = 0.4003 -------------+------------------------------ Adj R-squared = 0.3971 Total | 331293604 1150 288081.395 Root MSE = 416.75 -------------------------------------------------------------------------------- def_social_h~d | Coef. Std. Err. t P>|t| [95% Conf. Interval] ---------------+---------------------------------------------------------------- log_pctforeign | 220.8124 16.05465 13.75 0.000 189.3126 252.3123 | canton | 2 | -670.2634 67.5107 -9.93 0.000 -802.7221 -537.8047 3 | -94.19855 40.78477 -2.31 0.021 -174.2199 -14.17721 5 | 468.9531 35.77836 13.11 0.000 398.7545 539.1516 6 | -1193.182 86.82131 -13.74 0.000 -1363.529 -1022.835 7 | -468.4175 41.32189 -11.34 0.000 -549.4927 -387.3424 | _cons | 1180.505 40.99826 28.79 0.000 1100.065 1260.946 -------------------------------------------------------------------------------- Canton 1 is a ‘reference category’ – the intercept for canton 1 is “_cons”

  5. Dummy Variables This is not the same as running separate regressions for each canton, because we still assume that the slope is identical for every subgroup. Social Spending = B0 + B1*Log_Pctforeign + B2*Canton1 + B3*Canton2 + B4*Canton3 … For municipalities in Canton 1: Social Spending = B0 + B1*Log_Pctforeign + B2*1 + B3*0+ B4*0 … Social Spending = (B0 + B2) + B1*Log_Pctforeign For municipalities in Canton 2: Social Spending = B0 + B1*Log_Pctforeign + B2*0 + B3*1+ B4*0 … Social Spending = (B0 + B3) + B1*Log_Pctforeign All we are doing is changing the starting value -- allowing the expectation when log_pctforeign = 0 to differ across cantons. B1 still describes the effect of Log_Pctforeign on Social Spending across the entire sample.

  6. Dummy Variables What would happen if I added the following variables to the previous regression: • Variable that measures whether a canton is German-speaking (1) or French-speaking (0) • Variable that measures average GDP per capita in the canton. • Variable that measures whether a municipality allows (1) or does not allow (0) immigrants to vote.

  7. Dummy Variables and Interactions We can allow the relationship between log_pctforeign and def_social_head (the “slope”) to vary for each canton by using interactions: Social Spending = B0 + B1*Log_Pctforeign + B2*Canton1 + B3*Canton2 … + B4*Canton1*Log_Pctforeign+ B5*Canton2*Log_Pctforeign … For municipalities in Canton 1: Social Spending = B0 + B1*Log_Pctforeign + B2*1 + B3*0+ B4*1*Log_Pctforeign + B5*0*Log_Pctforeign Social Spending = (B0 + B2) + (B1+ B4)Log_Pctforeign For municipalities in Canton 2: Social Spending = (B0 + B3) + (B1 + B5)Log_Pctforeign This is the same as estimating the relationship between social spending and log_pctforeign separately for each subgroup. We are assuming that the relationship differs in each canton.

  8. Interactions Interactions should be justified by theory. For instance, we might reasonably assume that the relationship between % foreign and social expenditure would be different if municipalities allowed immigrants to vote. Voteright is a variable coded 1 if immigrants have voting rights in a municipality, 0 otherwise. . gen logpctforeignXvoteright = voteright * log_pctforeign . regdef_social_headlog_pctforeignlogpctforeignXvoterightvoterighti.canton if year == 2005 ----------------------------------------------------------------------------------------- def_social_head | Coef. Std. Err. t P>|t| [95% Conf. Interval] ------------------------+---------------------------------------------------------------- log_pctforeign | 179.9446 19.08151 9.43 0.000 142.5076 217.3816 logpctforeignXvoteright | 85.12434 32.40203 2.63 0.009 21.55309 148.6956 voteright | 112.3438 417.7021 0.27 0.788 -707.1681 931.8557 | canton | 2 | -703.456 68.08389 -10.33 0.000 -837.0333 -569.8786 3 | -73.26651 41.22884 -1.78 0.076 -154.1556 7.62253 4 | -1033.193 60.82019 -16.99 0.000 -1152.519 -913.8663 5 | 361.7032 411.5004 0.88 0.380 -445.6411 1169.048 6 | -1079.953 424.194 -2.55 0.011 -1912.201 -247.7039 7 | -559.3247 409.6522 -1.37 0.172 -1363.043 244.3936 | _cons | 1160.733 414.167 2.80 0.005 348.1573 1973.31 -----------------------------------------------------------------------------------------

  9. Interactions Social Spending = B0 + B1*Log_Pctforeign + B2*Canton1 + B3*Canton2 + … B4*VotingRights + B5*Log_Pctforeign*VotingRights For non-immigrant voting municipalities in Canton 1: Social Spending = B0 + B1*Log_Pctforeign + B2*1 + B3*0+ +B4*0 + B5*Log_Pctforeign*0 Social Spending = (B0+B2) + B1*Log_Pctforeign For immigrant voting municipalities in Canton 1: Social Spending = B0 + B1*Log_Pctforeign + B2*1 + B3*0+ +B4*1 + B5*Log_Pctforeign*1 Social Spending = (B0+B2+B4) + (B1+B5)*Log_Pctforeign

  10. Direct + Indirect Effects Direct Effect = Multivariate Regression Coefficient Indirect Effect = Bivariate – Multivariate

More Related