P-Value Interpretation for Dummy Variables (output included)

[Quinn Fraser]

Hello,

I am having some difficulty interpreting the p-value of my results - especially the dummy variables. Are you able to interpret the p-values of dummy variables? If not, how can I properly code my regression to incorporate the dummy variables? If you can interpret the p-value of dummy variables, what happens when some are statistically significant, and others aren't? Am I doing something wrong here?

Thanks in advance,

Quinn

---

Variable List:

Dependent variable: Public support for income equality
Independent variables: income, skilled labour, ethnicity
Control: education level, employement status, number of children, sex, age range

---

regress moreequal highincome i.moreskill b3.ethnicity i.highedu b5.employstat morechild i.female i.agerange


Source | SS df MS Number of obs = 2,616
-------------+---------------------------------- F(19, 2596) = 9.98
Model | 1181.16033 19 62.1663333 Prob > F = 0.0000
Residual | 16165.6287 2,596 6.22712968 R-squared = 0.0681
-------------+---------------------------------- Adj R-squared = 0.0613
Total | 17346.789 2,615 6.63357132 Root MSE = 2.4954

--------------------------------------------------------------------------------------
moreequal | Coef. Std. Err. t P>|t| [95% Conf. Interval]
---------------------+----------------------------------------------------------------
highincome | -.1642055 .0259355 -6.33 0.000 -.2150619 -.1133492

moreskill |
Semi-skilled | .4099856 .1308398 3.13 0.002 .1534246 .6665466
Skilled | -.485611 .1486537 -3.27 0.001 -.7771029 -.1941191

ethnicity |
Black | -.5094474 .1684931 -3.02 0.003 -.8398417 -.179053
Coloured | .2522887 .2262135 1.12 0.265 -.1912884 .6958658
Other | -.6747567 .3086933 -2.19 0.029 -1.280067 -.0694467

highedu |
Primary | -.3592873 .2671538 -1.34 0.179 -.8831433 .1645687
Secondary | -.5459883 .2726625 -2.00 0.045 -1.080646 -.0113303
Post-Secondary | -.7621883 .3413934 -2.23 0.026 -1.431619 -.0927573

employstat |
Full-time | -.1003902 .1315503 -0.76 0.445 -.3583444 .1575639
Part-time | -.3351507 .1939232 -1.73 0.084 -.7154105 .0451092
Self-employed | -.1412256 .2606941 -0.54 0.588 -.6524151 .3699639
Not in labour force | -.1179686 .1585718 -0.74 0.457 -.4289085 .1929714

morechild | -.1130585 .0492506 -2.30 0.022 -.2096329 -.0164842

female |
Female | .0241241 .1002043 0.24 0.810 -.1723645 .2206126

agerange |
30-39 | .1037347 .1343116 0.77 0.440 -.159634 .3671034
40-49 | .3456979 .1630838 2.12 0.034 .0259104 .6654854
50-69 | .1078065 .1914722 0.56 0.573 -.2676472 .4832601
60 and older | .1870045 .258449 0.72 0.469 -.3197825 .6937915

_cons | 6.716194 .363987 18.45 0.000 6.00246 7.429928

[Marcos Almeida]

The interpretation of of p-values for the predictors of a regression analysis is found galore in any decent introductory text on stats. Please do read the basics, otherwise it will be cumbersome to progress with the analysis.

Please read the FAQ as well. There you’ll find advice about sharing data/command/output in the forum.