- Country was added as an explanatory variable in the GLM procedure. Effective coding or parameterization was applied to the country attribute to create 6 explanatory categories and allow for group comparisons. The countries were categorized into 6 continental groups: Africa(0), Asia(1), Europe(2), North America(3), Oceania(4), South America(5).
- The quantitative explanatory variables were standardized for the GLM and Logistic procedures.
- The categorical response variable female employ was created from the quantitative variable as follows: Female employ rates < 50% = 0 and Female employ rates >= 50% = 1. This was used in the Logistic procedure.
Summary of Findings
- The GLM procedure shows the female employment rate is significantly correlated with the urban rate, income per person, polity score, life expectancy and employ rate. There is no correlation with internet use rate.
- When African countries are the reference explanatory group, only the European countries are statistically different.
- When Asian countries are the reference explanatory group, only the European countries are statistically different.
- When European countries are the reference explanatory group, all countries, except Oceanic countries, are statistically different.
- When North American countries are the reference explanatory group, only the European countries are statistically different.
- When Oceanic countries are the reference explanatory group, none of the other countries are statistically different.
- When South American countries are the reference explanatory group, only the European countries are statistically different.
- The Logistic procedure shows that countries with higher urban rates are have a 0.98 more likely to have female employ rates below 50% (parameter estimate = -.0164, p-value=0.0094, odds ratio=0.984). We can say with 95% confidence that the likelihood lies between 0.972 and 0.996.
- Adding income per person and controlling for urban rate shows no statistical significance. Adding internet use rate shows a similar result. However, this confounds the relationship between urban rate and the response category. Polity score and life expectancy also show statistical significance when controlling for the other explanatory variables.
- The employ rate explanatory variable (parameter estimate=0.3, p-value <0.0001, odds ratio=1.34) shows that countries with high employ rates are 1.34 times more likely to have female employ rates of 50%+. There is 95% confidence that the likelihood falls between 1.2 and 1.49.
The results support my original hypthesis of the inverse but significant relationship between the female employ rate and the urban rate.
SAS Code
PROC IMPORT DATAFILE="/home/mst07221/gapminder.csv"
DBMS=CSV
OUT=WORK.GAPMINDER;
GETNAMES=YES;
RUN;
******************************************************************************************
DATA MANAGEMENT
*******************************************************************************************;
*****************************************************
Parameterization of the country attribute:
0 = Africa
1 = Asia
2 = Europe
3 = North America
4 = Oceania
5 = South America
*****************************************************;
data new; set WORK.GAPMINDER;
if country = 'Algeria' then country_c = 0;
if country = 'Angola' then country_c = 0;
if country = 'Benin' then country_c = 0;
if country = 'Botswana' then country_c = 0;
if country = 'Burkina Faso' then country_c = 0;
if country = 'Burundi' then country_c = 0;
if country = 'Cameroon' then country_c = 0;
if country = 'Cape Verde' then country_c = 0;
if country = 'Central African Rep.' then country_c = 0;
if country = 'Chad' then country_c = 0;
if country = 'Comoros' then country_c = 0;
if country = 'Congo, Dem. Rep.' then country_c = 0;
if country = 'Congo, Rep.' then country_c = 0;
if country = 'Cote d''Ivoire' then country_c = 0;
if country = 'Djibouti' then country_c = 0;
if country = 'Egypt' then country_c = 0;
if country = 'Equatorial Guinea' then country_c = 0;
if country = 'Eritrea' then country_c = 0;
if country = 'Ethiopia' then country_c = 0;
if country = 'Gabon' then country_c = 0;
if country = 'Gambia' then country_c = 0;
if country = 'Ghana' then country_c = 0;
if country = 'Guinea' then country_c = 0;
if country = 'Guinea-Bissau' then country_c = 0;
if country = 'Kenya' then country_c = 0;
if country = 'Lesotho' then country_c = 0;
if country = 'Liberia' then country_c = 0;
if country = 'Libya' then country_c = 0;
if country = 'Madagascar' then country_c = 0;
if country = 'Malawi' then country_c = 0;
if country = 'Mali' then country_c = 0;
if country = 'Mauritania' then country_c = 0;
if country = 'Mauritius' then country_c = 0;
if country = 'Morocco' then country_c = 0;
if country = 'Mozambique' then country_c = 0;
if country = 'Namibia' then country_c = 0;
if country = 'Niger' then country_c = 0;
if country = 'Nigeria' then country_c = 0;
if country = 'Reunion' then country_c = 0;
if country = 'Rwanda' then country_c = 0;
if country = 'Sao Tome and Principe' then country_c = 0;
if country = 'Senegal' then country_c = 0;
if country = 'Seychelles' then country_c = 0;
if country = 'Sierra Leone' then country_c = 0;
if country = 'Somalia' then country_c = 0;
if country = 'South Africa' then country_c = 0;
if country = 'Sudan' then country_c = 0;
if country = 'Swaziland' then country_c = 0;
if country = 'Tanzania' then country_c = 0;
if country = 'Togo' then country_c = 0;
if country = 'Tunisia' then country_c = 0;
if country = 'Uganda' then country_c = 0;
if country = 'Zambia' then country_c = 0;
if country = 'Zimbabwe' then country_c = 0;
if country = 'Afghanistan' then country_c = 1;
if country = 'Bahrain' then country_c = 1;
if country = 'Bangladesh' then country_c = 1;
if country = 'Bhutan' then country_c = 1;
if country = 'Brunei' then country_c = 1;
if country = 'Cambodia' then country_c = 1;
if country = 'China' then country_c = 1;
if country = 'Guam' then country_c = 1;
if country = 'Hong Kong, China' then country_c = 1;
if country = 'India' then country_c = 1;
if country = 'Indonesia' then country_c = 1;
if country = 'Iran' then country_c = 1;
if country = 'Iraq' then country_c = 1;
if country = 'Israel' then country_c = 1;
if country = 'Japan' then country_c = 1;
if country = 'Jordan' then country_c = 1;
if country = 'Kazakhstan' then country_c = 1;
if country = 'Korea, Dem. Rep.' then country_c = 1;
if country = 'Korea, Rep.' then country_c = 1;
if country = 'Kuwait' then country_c = 1;
if country = 'Kyrgyzstan' then country_c = 1;
if country = 'Laos' then country_c = 1;
if country = 'Lebanon' then country_c = 1;
if country = 'Macao, China' then country_c = 1;
if country = 'Malaysia' then country_c = 1;
if country = 'Maldives' then country_c = 1;
if country = 'Mongolia' then country_c = 1;
if country = 'Myanmar' then country_c = 1;
if country = 'Nepal' then country_c = 1;
if country = 'Oman' then country_c = 1;
if country = 'Pakistan' then country_c = 1;
if country = 'Philippines' then country_c = 1;
if country = 'Qatar' then country_c = 1;
if country = 'Russia' then country_c = 1;
if country = 'Saudi Arabia' then country_c = 1;
if country = 'Singapore' then country_c = 1;
if country = 'Sri Lanka' then country_c = 1;
if country = 'Syria' then country_c = 1;
if country = 'Taiwan' then country_c = 1;
if country = 'Tajikistan' then country_c = 1;
if country = 'Thailand' then country_c = 1;
if country = 'Timor-Leste' then country_c = 1;
if country = 'Turkey' then country_c = 1;
if country = 'Turkmenistan' then country_c = 1;
if country = 'United Arab Emirates' then country_c = 1;
if country = 'Uzbekistan' then country_c = 1;
if country = 'Vietnam' then country_c = 1;
if country = 'West Bank and Gaza' then country_c = 1;
if country = 'Yemen, Rep.' then country_c = 1;
if country = 'Albania' then country_c = 2;
if country = 'Andorra' then country_c = 2;
if country = 'Armenia' then country_c = 2;
if country = 'Austria' then country_c = 2;
if country = 'Azerbaijan' then country_c = 2;
if country = 'Belarus' then country_c = 2;
if country = 'Belgium' then country_c = 2;
if country = 'Bosnia and Herzegovina' then country_c = 2;
if country = 'Bulgaria' then country_c = 2;
if country = 'Croatia' then country_c = 2;
if country = 'Cyprus' then country_c = 2;
if country = 'Czech Rep.' then country_c = 2;
if country = 'Denmark' then country_c = 2;
if country = 'Estonia' then country_c = 2;
if country = 'Faeroe Islands' then country_c = 2;
if country = 'Finland' then country_c = 2;
if country = 'France' then country_c = 2;
if country = 'Georgia' then country_c = 2;
if country = 'Germany' then country_c = 2;
if country = 'Gibraltar' then country_c = 2;
if country = 'Greece' then country_c = 2;
if country = 'Hungary' then country_c = 2;
if country = 'Iceland' then country_c = 2;
if country = 'Ireland' then country_c = 2;
if country = 'Italy' then country_c = 2;
if country = 'Latvia' then country_c = 2;
if country = 'Liechtenstein' then country_c = 2;
if country = 'Lithuania' then country_c = 2;
if country = 'Luxembourg' then country_c = 2;
if country = 'Macedonia, FYR' then country_c = 2;
if country = 'Malta' then country_c = 2;
if country = 'Moldova' then country_c = 2;
if country = 'Monaco' then country_c = 2;
if country = 'Montenegro' then country_c = 2;
if country = 'Netherlands' then country_c = 2;
if country = 'Norway' then country_c = 2;
if country = 'Poland' then country_c = 2;
if country = 'Portugal' then country_c = 2;
if country = 'Romania' then country_c = 2;
if country = 'San Marino' then country_c = 2;
if country = 'Serbia' then country_c = 2;
if country = 'Serbia and Montenegro' then country_c = 2;
if country = 'Slovak Republic' then country_c = 2;
if country = 'Slovenia' then country_c = 2;
if country = 'Spain' then country_c = 2;
if country = 'Sweden' then country_c = 2;
if country = 'Switzerland' then country_c = 2;
if country = 'Ukraine' then country_c = 2;
if country = 'United Kingdom' then country_c = 2;
if country = 'Antigua and Barbuda' then country_c = 3;
if country = 'Aruba' then country_c = 3;
if country = 'Bahamas' then country_c = 3;
if country = 'Barbados' then country_c = 3;
if country = 'Belize' then country_c = 3;
if country = 'Bermuda' then country_c = 3;
if country = 'Canada' then country_c = 3;
if country = 'Cayman Islands' then country_c = 3;
if country = 'Cook Islands' then country_c = 3;
if country = 'Costa Rica' then country_c = 3;
if country = 'Cuba' then country_c = 3;
if country = 'Dominica' then country_c = 3;
if country = 'Dominican Rep.' then country_c = 3;
if country = 'El Salvador' then country_c = 3;
if country = 'Greenland' then country_c = 3;
if country = 'Grenada' then country_c = 3;
if country = 'Guadeloupe' then country_c = 3;
if country = 'Guatemala' then country_c = 3;
if country = 'Haiti' then country_c = 3;
if country = 'Honduras' then country_c = 3;
if country = 'Jamaica' then country_c = 3;
if country = 'Martinique' then country_c = 3;
if country = 'Mexico' then country_c = 3;
if country = 'Netherlands Antilles' then country_c = 3;
if country = 'Nicaragua' then country_c = 3;
if country = 'Panama' then country_c = 3;
if country = 'Puerto Rico' then country_c = 3;
if country = 'Saint Kitts and Nevis' then country_c = 3;
if country = 'Saint Lucia' then country_c = 3;
if country = 'Saint Vincent and the Grenadines' then country_c = 3;
if country = 'Trinidad and Tobago' then country_c = 3;
if country = 'United States' then country_c = 3;
if country = 'Australia' then country_c = 4;
if country = 'Fiji' then country_c = 4;
if country = 'French Polynesia' then country_c = 4;
if country = 'Kiribati' then country_c = 4;
if country = 'Marshall Islands' then country_c = 4;
if country = 'Micronesia, Fed. Sts.' then country_c = 4;
if country = 'Nauru' then country_c = 4;
if country = 'New Caledonia' then country_c = 4;
if country = 'New Zealand' then country_c = 4;
if country = 'Niue' then country_c = 4;
if country = 'Palau' then country_c = 4;
if country = 'Papua New Guinea' then country_c = 4;
if country = 'Samoa' then country_c = 4;
if country = 'Solomon Islands' then country_c = 4;
if country = 'Tonga' then country_c = 4;
if country = 'Tuvalu' then country_c = 4;
if country = 'Vanuatu' then country_c = 4;
if country = 'Argentina' then country_c = 5;
if country = 'Bolivia' then country_c = 5;
if country = 'Brazil' then country_c = 5;
if country = 'Chile' then country_c = 5;
if country = 'Colombia' then country_c = 5;
if country = 'Ecuador' then country_c = 5;
if country = 'Guyana' then country_c = 5;
if country = 'Paraguay' then country_c = 5;
if country = 'Peru' then country_c = 5;
if country = 'Suriname' then country_c = 5;
if country = 'Uruguay' then country_c = 5;
if country = 'Venezuela' then country_c = 5;
**make the quantitative response variable a binary variable
** split the female employment rate to above(1)/below(0) 50%;
if femaleemployrate < 50 then fememploy = 0;
else fememploy = 1;
run;
**************************************************************************************
END DATA MANAGEMENT
**************************************************************************************;
**************************************************************************************
CATEGORICAL EXPLANATORY VARIABLES RE-VISITED (3+ CATEGORIES)
***************************************************************************************;
* centering quantitative explanatory variables;
* print mean;
PROC MEANS;
var femaleemployrate urbanrate incomeperperson internetuserate polityscore lifeexpectancy employrate;
run;
PROC GLM;
model femaleemployrate=urbanrate incomeperperson internetuserate polityscore lifeexpectancy employrate/solution;
run;
* centering (subtract mean);
data new2;
set new;
urbanrate_c = urbanrate - 56.7693596;
incomeperperson_c = incomeperperson - 8740.97;
internetuserate_c = internetuserate - 35.6327158;
polityscore_c = polityscore - 3.6894410;
lifeexpectancy_c = lifeexpectancy - 69.7535236;
employrate_c = employrate - 58.6359551;
run;
* check coding;
PROC MEANS;
var urbanrate_c incomeperperson_c internetuserate_c polityscore_c lifeexpectancy_c employrate_c;
run;
** simple linear regression modelling;
PROC GLM;
model femaleemployrate=urbanrate_c incomeperperson_c internetuserate_c polityscore_c lifeexpectancy_c employrate_c/solution;
run;
*Reference group = countries in Africa;
PROC GLM;
class country_c (ref="0");
model femaleemployrate=urbanrate_c incomeperperson_c internaeeserate_c polityscore_c
lifeexpectancy_c employrate_ccountry_c/solution;
run;
* Reference group = countries in Asia;
PROC GLM;
class country_c (ref="1");
model femaleemployrate=urbanrate_c incomeperperson_c internaeeserate_c polityscore_c
lifeexpectancy_c employrate_ccountry_c/solution;
run;
* Reference group = countries in Europe;
PROC GLM;
class country_c (ref="2");
model femaleemployrate=urbanrate_c incomeperperson_c internaeeserate_c polityscore_c
lifeexpectancy_c employrate_ccountry_c/solution;
run;
* Reference group = countries in North America;
PROC GLM;
class country_c (ref="3");
model femaleemployrate=urbanrate_c incomeperperson_c internaeeserate_c polityscore_c
lifeexpectancy_c employrate_ccountry_c/solution;
run;
* Reference group = countries in Oceania;
PROC GLM;
class country_c (ref="4");
model femaleemployrate=urbanrate_c incomeperperson_c internaeeserate_c polityscore_c
lifeexpectancy_c employrate_ccountry_c/solution;
run;
* Reference group = countries in South America;
PROC GLM;
class country_c (ref="5");
model femaleemployrate=urbanrate_c incomeperperson_c internaeeserate_c polityscore_c
lifeexpectancy_c employrate_ccountry_c/solution;
run;
*****************************************************************************************
LOGISTIC REGRESSION
*****************************************************************************************;
Proc logistic descending;
model fememploy= urbanrate;_c
run;
Proc logistic descending;
model fememploy= urbanrate_c; incomeperperson_c
run;
Proc logistic descending;
modee fememploy= urbanrate _c incomeperperson_c internetuserate_c
run;
PProc logistic descending;
model fememploy= urbanrate_c incomeperperson_c internetuserate_c polityscore_c;
run;
Proc logistic descending;
model fememploy= urbanrate_c incomeperperson_c internetuserate_c polityscore_c;
run;
Proc logistic descending;
model fememploy= urbanrate_c incomeperperson_c internetuserate_c polityscore_c lifeexpectancy_c;
run;
Proc logistic descending;
model fememploy= urbanrate_c incomeperperson_c internetuserate_c polityscore_c lifeexpectancy_c employrate_c;
run;
Results
The MEANS Procedure
The MEANS Procedure
| Variable | N | Mean | Std Dev | Minimum | Maximum |
|---|---|---|---|---|---|
femaleemployrate
urbanrate
incomeperperson
internetuserate
polityscore
lifeexpectancy
employrate
|
178
203
190
192
161
191
178
|
47.5494381
56.7693596
8740.97
35.6327158
3.6894410
69.7535236
58.6359551
|
14.6257425
23.8449326
14262.81
27.7802846
6.3148991
9.7086205
10.5194545
|
11.3000002
10.4000000
103.7758572
0.2100663
-10.0000000
47.7940000
32.0000000
|
83.3000031
100.0000000
105147.44
95.6381132
10.0000000
83.3940000
83.1999969
|
The GLM Procedure
| Number of Observations Read | 213 |
|---|---|
| Number of Observations Used | 150 |
The GLM Procedure
Dependent Variable: femaleemployrate
| Source | DF | Sum of Squares | Mean Square | F Value | Pr > F |
|---|---|---|---|---|---|
| Model | 6 | 25581.91334 | 4263.65222 | 87.46 | <.0001 |
| Error | 143 | 6971.29659 | 48.75033 | ||
| Corrected Total | 149 | 32553.20994 |
| R-Square | Coeff Var | Root MSE | femaleemployrate Mean |
|---|---|---|---|
| 0.785849 | 14.51569 | 6.982143 | 48.10067 |
| Source | DF | Type I SS | Mean Square | F Value | Pr > F |
|---|---|---|---|---|---|
| urbanrate | 1 | 4068.92866 | 4068.92866 | 83.46 | <.0001 |
| incomeperperson | 1 | 1806.33239 | 1806.33239 | 37.05 | <.0001 |
| internetuserate | 1 | 18.84957 | 18.84957 | 0.39 | 0.5351 |
| polityscore | 1 | 196.33302 | 196.33302 | 4.03 | 0.0467 |
| lifeexpectancy | 1 | 1716.20313 | 1716.20313 | 35.20 | <.0001 |
| employrate | 1 | 17775.26657 | 17775.26657 | 364.62 | <.0001 |
| Source | DF | Type III SS | Mean Square | F Value | Pr > F |
|---|---|---|---|---|---|
| urbanrate | 1 | 161.04308 | 161.04308 | 3.30 | 0.0712 |
| incomeperperson | 1 | 56.36899 | 56.36899 | 1.16 | 0.2840 |
| internetuserate | 1 | 398.94589 | 398.94589 | 8.18 | 0.0049 |
| polityscore | 1 | 835.38354 | 835.38354 | 17.14 | <.0001 |
| lifeexpectancy | 1 | 284.70984 | 284.70984 | 5.84 | 0.0169 |
| employrate | 1 | 17775.26657 | 17775.26657 | 364.62 | <.0001 |
| Parameter | Estimate | Standard Error | t Value | Pr > |t| |
|---|---|---|---|---|
| Intercept | -8.394978585 | 7.72729710 | -1.09 | 0.2791 |
| urbanrate | -0.069392059 | 0.03817926 | -1.82 | 0.0712 |
| incomeperperson | -0.000112492 | 0.00010461 | -1.08 | 0.2840 |
| internetuserate | 0.132279749 | 0.04624078 | 2.86 | 0.0049 |
| polityscore | 0.411590133 | 0.09942847 | 4.14 | <.0001 |
| lifeexpectancy | -0.233124441 | 0.09646624 | -2.42 | 0.0169 |
| employrate | 1.202132527 | 0.06295541 | 19.09 | <.0001 |
The MEANS Procedure
| Variable | N | Mean | Std Dev | Minimum | Maximum |
|---|---|---|---|---|---|
urbanrate_c
incomeperperson_c
internetuserate_c
polityscore_c
lifeexpectancy_c
employrate_c
|
203
190
192
161
191
178
|
5.9113279E-9
-0.0039237
-3.097827E-8
-6.211181E-9
-3.979057E-8
-1.810316E-8
|
23.8449326
14262.81
27.7802846
6.3148991
9.7086205
10.5194545
|
-46.3693596
-8637.19
-35.4226495
-13.6894410
-21.9595236
-26.6359551
|
43.2306404
96406.47
60.0053974
6.3105590
13.6404764
24.5640418
|
The GLM Procedure
| Number of Observations Read | 213 |
|---|---|
| Number of Observations Used | 150 |
The GLM Procedure
Dependent Variable: femaleemployrate
| Source | DF | Sum of Squares | Mean Square | F Value | Pr > F |
|---|---|---|---|---|---|
| Model | 6 | 25581.91334 | 4263.65222 | 87.46 | <.0001 |
| Error | 143 | 6971.29659 | 48.75033 | ||
| Corrected Total | 149 | 32553.20994 |
| R-Square | Coeff Var | Root MSE | femaleemployrate Mean |
|---|---|---|---|
| 0.785849 | 14.51569 | 6.982143 | 48.10067 |
| Source | DF | Type I SS | Mean Square | F Value | Pr > F |
|---|---|---|---|---|---|
| urbanrate_c | 1 | 4068.92866 | 4068.92866 | 83.46 | <.0001 |
| incomeperperson_c | 1 | 1806.33239 | 1806.33239 | 37.05 | <.0001 |
| internetuserate_c | 1 | 18.84957 | 18.84957 | 0.39 | 0.5351 |
| polityscore_c | 1 | 196.33302 | 196.33302 | 4.03 | 0.0467 |
| lifeexpectancy_c | 1 | 1716.20313 | 1716.20313 | 35.20 | <.0001 |
| employrate_c | 1 | 17775.26657 | 17775.26657 | 364.62 | <.0001 |
| Source | DF | Type III SS | Mean Square | F Value | Pr > F |
|---|---|---|---|---|---|
| urbanrate_c | 1 | 161.04308 | 161.04308 | 3.30 | 0.0712 |
| incomeperperson_c | 1 | 56.36899 | 56.36899 | 1.16 | 0.2840 |
| internetuserate_c | 1 | 398.94589 | 398.94589 | 8.18 | 0.0049 |
| polityscore_c | 1 | 835.38354 | 835.38354 | 17.14 | <.0001 |
| lifeexpectancy_c | 1 | 284.70984 | 284.70984 | 5.84 | 0.0169 |
| employrate_c | 1 | 17775.26657 | 17775.26657 | 364.62 | <.0001 |
| Parameter | Estimate | Standard Error | t Value | Pr > |t| |
|---|---|---|---|---|
| Intercept | 47.14134716 | 0.58829818 | 80.13 | <.0001 |
| urbanrate_c | -0.06939206 | 0.03817926 | -1.82 | 0.0712 |
| incomeperperson_c | -0.00011249 | 0.00010461 | -1.08 | 0.2840 |
| internetuserate_c | 0.13227975 | 0.04624078 | 2.86 | 0.0049 |
| polityscore_c | 0.41159013 | 0.09942847 | 4.14 | <.0001 |
| lifeexpectancy_c | -0.23312444 | 0.09646624 | -2.42 | 0.0169 |
| employrate_c | 1.20213253 | 0.06295541 | 19.09 | <.0001 |
The GLM Procedure
| Class Level Information | ||
|---|---|---|
| Class | Levels | Values |
| country_c | 6 | 1 2 3 4 5 0 |
| Number of Observations Read | 213 |
|---|---|
| Number of Observations Used | 148 |
The GLM Procedure
Dependent Variable: femaleemployrate
| Source | DF | Sum of Squares | Mean Square | F Value | Pr > F |
|---|---|---|---|---|---|
| Model | 11 | 26988.92492 | 2453.53863 | 63.82 | <.0001 |
| Error | 136 | 5228.50225 | 38.44487 | ||
| Corrected Total | 147 | 32217.42717 |
| R-Square | Coeff Var | Root MSE | femaleemployrate Mean |
|---|---|---|---|
| 0.837712 | 12.89770 | 6.200393 | 48.07365 |
| Source | DF | Type I SS | Mean Square | F Value | Pr > F |
|---|---|---|---|---|---|
| urbanrate_c | 1 | 3885.57081 | 3885.57081 | 101.07 | <.0001 |
| incomeperperson_c | 1 | 1885.43020 | 1885.43020 | 49.04 | <.0001 |
| internetuserate_c | 1 | 8.39997 | 8.39997 | 0.22 | 0.6409 |
| polityscore_c | 1 | 161.21451 | 161.21451 | 4.19 | 0.0425 |
| lifeexpectancy_c | 1 | 1624.64281 | 1624.64281 | 42.26 | <.0001 |
| employrate_c | 1 | 18435.23349 | 18435.23349 | 479.52 | <.0001 |
| country_c | 5 | 988.43314 | 197.68663 | 5.14 | 0.0002 |
| Source | DF | Type III SS | Mean Square | F Value | Pr > F |
|---|---|---|---|---|---|
| urbanrate_c | 1 | 25.07077 | 25.07077 | 0.65 | 0.4208 |
| incomeperperson_c | 1 | 7.92454 | 7.92454 | 0.21 | 0.6505 |
| internetuserate_c | 1 | 18.45581 | 18.45581 | 0.48 | 0.4896 |
| polityscore_c | 1 | 216.34246 | 216.34246 | 5.63 | 0.0191 |
| lifeexpectancy_c | 1 | 108.05325 | 108.05325 | 2.81 | 0.0959 |
| employrate_c | 1 | 17361.33158 | 17361.33158 | 451.59 | <.0001 |
| country_c | 5 | 988.43314 | 197.68663 | 5.14 | 0.0002 |
| Parameter | Estimate | Standard Error | t Value | Pr > |t| | |
|---|---|---|---|---|---|
| Intercept | 45.51220706 | B | 1.46691490 | 31.03 | <.0001 |
| urbanrate_c | -0.03051835 | 0.03779169 | -0.81 | 0.4208 | |
| incomeperperson_c | -0.00004507 | 0.00009926 | -0.45 | 0.6505 | |
| internetuserate_c | 0.03357976 | 0.04846522 | 0.69 | 0.4896 | |
| polityscore_c | 0.23990981 | 0.10113380 | 2.37 | 0.0191 | |
| lifeexpectancy_c | -0.19790675 | 0.11804861 | -1.68 | 0.0959 | |
| employrate_c | 1.29068372 | 0.06073619 | 21.25 | <.0001 | |
| country_c 1 | -0.65423435 | B | 1.90496373 | -0.34 | 0.7318 |
| country_c 2 | 7.96674783 | B | 2.49318124 | 3.20 | 0.0017 |
| country_c 3 | -1.55741061 | B | 2.54395495 | -0.61 | 0.5414 |
| country_c 4 | 1.90130435 | B | 3.28376933 | 0.58 | 0.5635 |
| country_c 5 | 0.74774921 | B | 2.82427232 | 0.26 | 0.7916 |
| country_c 0 | 0.00000000 | B | . | . | . |
The X'X matrix has been found to be singular, and a generalized inverse was used to solve the normal equations. Terms whose estimates are followed by the letter 'B' are not uniquely estimable.
The GLM Procedure
| Class Level Information | ||
|---|---|---|
| Class | Levels | Values |
| country_c | 6 | 0 2 3 4 5 1 |
| Number of Observations Read | 213 |
|---|---|
| Number of Observations Used | 148 |
The GLM Procedure
Dependent Variable: femaleemployrate
| Source | DF | Sum of Squares | Mean Square | F Value | Pr > F |
|---|---|---|---|---|---|
| Model | 11 | 26988.92492 | 2453.53863 | 63.82 | <.0001 |
| Error | 136 | 5228.50225 | 38.44487 | ||
| Corrected Total | 147 | 32217.42717 |
| R-Square | Coeff Var | Root MSE | femaleemployrate Mean |
|---|---|---|---|
| 0.837712 | 12.89770 | 6.200393 | 48.07365 |
| Source | DF | Type I SS | Mean Square | F Value | Pr > F |
|---|---|---|---|---|---|
| urbanrate_c | 1 | 3885.57081 | 3885.57081 | 101.07 | <.0001 |
| incomeperperson_c | 1 | 1885.43020 | 1885.43020 | 49.04 | <.0001 |
| internetuserate_c | 1 | 8.39997 | 8.39997 | 0.22 | 0.6409 |
| polityscore_c | 1 | 161.21451 | 161.21451 | 4.19 | 0.0425 |
| lifeexpectancy_c | 1 | 1624.64281 | 1624.64281 | 42.26 | <.0001 |
| employrate_c | 1 | 18435.23349 | 18435.23349 | 479.52 | <.0001 |
| country_c | 5 | 988.43314 | 197.68663 | 5.14 | 0.0002 |
| Source | DF | Type III SS | Mean Square | F Value | Pr > F |
|---|---|---|---|---|---|
| urbanrate_c | 1 | 25.07077 | 25.07077 | 0.65 | 0.4208 |
| incomeperperson_c | 1 | 7.92454 | 7.92454 | 0.21 | 0.6505 |
| internetuserate_c | 1 | 18.45581 | 18.45581 | 0.48 | 0.4896 |
| polityscore_c | 1 | 216.34246 | 216.34246 | 5.63 | 0.0191 |
| lifeexpectancy_c | 1 | 108.05325 | 108.05325 | 2.81 | 0.0959 |
| employrate_c | 1 | 17361.33158 | 17361.33158 | 451.59 | <.0001 |
| country_c | 5 | 988.43314 | 197.68663 | 5.14 | 0.0002 |
| Parameter | Estimate | Standard Error | t Value | Pr > |t| | |
|---|---|---|---|---|---|
| Intercept | 44.85797271 | B | 1.11704976 | 40.16 | <.0001 |
| urbanrate_c | -0.03051835 | 0.03779169 | -0.81 | 0.4208 | |
| incomeperperson_c | -0.00004507 | 0.00009926 | -0.45 | 0.6505 | |
| internetuserate_c | 0.03357976 | 0.04846522 | 0.69 | 0.4896 | |
| polityscore_c | 0.23990981 | 0.10113380 | 2.37 | 0.0191 | |
| lifeexpectancy_c | -0.19790675 | 0.11804861 | -1.68 | 0.0959 | |
| employrate_c | 1.29068372 | 0.06073619 | 21.25 | <.0001 | |
| country_c 0 | 0.65423435 | B | 1.90496373 | 0.34 | 0.7318 |
| country_c 2 | 8.62098218 | B | 1.90472422 | 4.53 | <.0001 |
| country_c 3 | -0.90317626 | B | 2.10153887 | -0.43 | 0.6680 |
| country_c 4 | 2.55553870 | B | 3.01657031 | 0.85 | 0.3984 |
| country_c 5 | 1.40198356 | B | 2.46450008 | 0.57 | 0.5704 |
| country_c 1 | 0.00000000 | B | . | . | . |
The X'X matrix has been found to be singular, and a generalized inverse was used to solve the normal equations. Terms whose estimates are followed by the letter 'B' are not uniquely estimable.
The GLM Procedure
| Class Level Information | ||
|---|---|---|
| Class | Levels | Values |
| country_c | 6 | 0 1 3 4 5 2 |
| Number of Observations Read | 213 |
|---|---|
| Number of Observations Used | 148 |
The GLM Procedure
Dependent Variable: femaleemployrate
| Source | DF | Sum of Squares | Mean Square | F Value | Pr > F |
|---|---|---|---|---|---|
| Model | 11 | 26988.92492 | 2453.53863 | 63.82 | <.0001 |
| Error | 136 | 5228.50225 | 38.44487 | ||
| Corrected Total | 147 | 32217.42717 |
| R-Square | Coeff Var | Root MSE | femaleemployrate Mean |
|---|---|---|---|
| 0.837712 | 12.89770 | 6.200393 | 48.07365 |
| Source | DF | Type I SS | Mean Square | F Value | Pr > F |
|---|---|---|---|---|---|
| urbanrate_c | 1 | 3885.57081 | 3885.57081 | 101.07 | <.0001 |
| incomeperperson_c | 1 | 1885.43020 | 1885.43020 | 49.04 | <.0001 |
| internetuserate_c | 1 | 8.39997 | 8.39997 | 0.22 | 0.6409 |
| polityscore_c | 1 | 161.21451 | 161.21451 | 4.19 | 0.0425 |
| lifeexpectancy_c | 1 | 1624.64281 | 1624.64281 | 42.26 | <.0001 |
| employrate_c | 1 | 18435.23349 | 18435.23349 | 479.52 | <.0001 |
| country_c | 5 | 988.43314 | 197.68663 | 5.14 | 0.0002 |
| Source | DF | Type III SS | Mean Square | F Value | Pr > F |
|---|---|---|---|---|---|
| urbanrate_c | 1 | 25.07077 | 25.07077 | 0.65 | 0.4208 |
| incomeperperson_c | 1 | 7.92454 | 7.92454 | 0.21 | 0.6505 |
| internetuserate_c | 1 | 18.45581 | 18.45581 | 0.48 | 0.4896 |
| polityscore_c | 1 | 216.34246 | 216.34246 | 5.63 | 0.0191 |
| lifeexpectancy_c | 1 | 108.05325 | 108.05325 | 2.81 | 0.0959 |
| employrate_c | 1 | 17361.33158 | 17361.33158 | 451.59 | <.0001 |
| country_c | 5 | 988.43314 | 197.68663 | 5.14 | 0.0002 |
| Parameter | Estimate | Standard Error | t Value | Pr > |t| | |
|---|---|---|---|---|---|
| Intercept | 53.47895489 | B | 1.46409556 | 36.53 | <.0001 |
| urbanrate_c | -0.03051835 | 0.03779169 | -0.81 | 0.4208 | |
| incomeperperson_c | -0.00004507 | 0.00009926 | -0.45 | 0.6505 | |
| internetuserate_c | 0.03357976 | 0.04846522 | 0.69 | 0.4896 | |
| polityscore_c | 0.23990981 | 0.10113380 | 2.37 | 0.0191 | |
| lifeexpectancy_c | -0.19790675 | 0.11804861 | -1.68 | 0.0959 | |
| employrate_c | 1.29068372 | 0.06073619 | 21.25 | <.0001 | |
| country_c 0 | -7.96674783 | B | 2.49318124 | -3.20 | 0.0017 |
| country_c 1 | -8.62098218 | B | 1.90472422 | -4.53 | <.0001 |
| country_c 3 | -9.52415844 | B | 2.25625098 | -4.22 | <.0001 |
| country_c 4 | -6.06544348 | B | 3.12448527 | -1.94 | 0.0543 |
| country_c 5 | -7.21899862 | B | 2.59052312 | -2.79 | 0.0061 |
| country_c 2 | 0.00000000 | B | . | . | . |
The X'X matrix has been found to be singular, and a generalized inverse was used to solve the normal equations. Terms whose estimates are followed by the letter 'B' are not uniquely estimable.
The GLM Procedure
| Class Level Information | ||
|---|---|---|
| Class | Levels | Values |
| country_c | 6 | 0 1 2 4 5 3 |
| Number of Observations Read | 213 |
|---|---|
| Number of Observations Used | 148 |
The GLM Procedure
Dependent Variable: femaleemployrate
| Source | DF | Sum of Squares | Mean Square | F Value | Pr > F |
|---|---|---|---|---|---|
| Model | 11 | 26988.92492 | 2453.53863 | 63.82 | <.0001 |
| Error | 136 | 5228.50225 | 38.44487 | ||
| Corrected Total | 147 | 32217.42717 |
| R-Square | Coeff Var | Root MSE | femaleemployrate Mean |
|---|---|---|---|
| 0.837712 | 12.89770 | 6.200393 | 48.07365 |
| Source | DF | Type I SS | Mean Square | F Value | Pr > F |
|---|---|---|---|---|---|
| urbanrate_c | 1 | 3885.57081 | 3885.57081 | 101.07 | <.0001 |
| incomeperperson_c | 1 | 1885.43020 | 1885.43020 | 49.04 | <.0001 |
| internetuserate_c | 1 | 8.39997 | 8.39997 | 0.22 | 0.6409 |
| polityscore_c | 1 | 161.21451 | 161.21451 | 4.19 | 0.0425 |
| lifeexpectancy_c | 1 | 1624.64281 | 1624.64281 | 42.26 | <.0001 |
| employrate_c | 1 | 18435.23349 | 18435.23349 | 479.52 | <.0001 |
| country_c | 5 | 988.43314 | 197.68663 | 5.14 | 0.0002 |
| Source | DF | Type III SS | Mean Square | F Value | Pr > F |
|---|---|---|---|---|---|
| urbanrate_c | 1 | 25.07077 | 25.07077 | 0.65 | 0.4208 |
| incomeperperson_c | 1 | 7.92454 | 7.92454 | 0.21 | 0.6505 |
| internetuserate_c | 1 | 18.45581 | 18.45581 | 0.48 | 0.4896 |
| polityscore_c | 1 | 216.34246 | 216.34246 | 5.63 | 0.0191 |
| lifeexpectancy_c | 1 | 108.05325 | 108.05325 | 2.81 | 0.0959 |
| employrate_c | 1 | 17361.33158 | 17361.33158 | 451.59 | <.0001 |
| country_c | 5 | 988.43314 | 197.68663 | 5.14 | 0.0002 |
| Parameter | Estimate | Standard Error | t Value | Pr > |t| | |
|---|---|---|---|---|---|
| Intercept | 43.95479645 | B | 1.82812622 | 24.04 | <.0001 |
| urbanrate_c | -0.03051835 | 0.03779169 | -0.81 | 0.4208 | |
| incomeperperson_c | -0.00004507 | 0.00009926 | -0.45 | 0.6505 | |
| internetuserate_c | 0.03357976 | 0.04846522 | 0.69 | 0.4896 | |
| polityscore_c | 0.23990981 | 0.10113380 | 2.37 | 0.0191 | |
| lifeexpectancy_c | -0.19790675 | 0.11804861 | -1.68 | 0.0959 | |
| employrate_c | 1.29068372 | 0.06073619 | 21.25 | <.0001 | |
| country_c 0 | 1.55741061 | B | 2.54395495 | 0.61 | 0.5414 |
| country_c 1 | 0.90317626 | B | 2.10153887 | 0.43 | 0.6680 |
| country_c 2 | 9.52415844 | B | 2.25625098 | 4.22 | <.0001 |
| country_c 4 | 3.45871496 | B | 3.27091579 | 1.06 | 0.2922 |
| country_c 5 | 2.30515982 | B | 2.62103664 | 0.88 | 0.3807 |
| country_c 3 | 0.00000000 | B | . | . | . |
The X'X matrix has been found to be singular, and a generalized inverse was used to solve the normal equations. Terms whose estimates are followed by the letter 'B' are not uniquely estimable.
The GLM Procedure
| Class Level Information | ||
|---|---|---|
| Class | Levels | Values |
| country_c | 6 | 0 1 2 3 5 4 |
| Number of Observations Read | 213 |
|---|---|
| Number of Observations Used | 148 |
The GLM Procedure
Dependent Variable: femaleemployrate
| Source | DF | Sum of Squares | Mean Square | F Value | Pr > F |
|---|---|---|---|---|---|
| Model | 11 | 26988.92492 | 2453.53863 | 63.82 | <.0001 |
| Error | 136 | 5228.50225 | 38.44487 | ||
| Corrected Total | 147 | 32217.42717 |
| R-Square | Coeff Var | Root MSE | femaleemployrate Mean |
|---|---|---|---|
| 0.837712 | 12.89770 | 6.200393 | 48.07365 |
| Source | DF | Type I SS | Mean Square | F Value | Pr > F |
|---|---|---|---|---|---|
| urbanrate_c | 1 | 3885.57081 | 3885.57081 | 101.07 | <.0001 |
| incomeperperson_c | 1 | 1885.43020 | 1885.43020 | 49.04 | <.0001 |
| internetuserate_c | 1 | 8.39997 | 8.39997 | 0.22 | 0.6409 |
| polityscore_c | 1 | 161.21451 | 161.21451 | 4.19 | 0.0425 |
| lifeexpectancy_c | 1 | 1624.64281 | 1624.64281 | 42.26 | <.0001 |
| employrate_c | 1 | 18435.23349 | 18435.23349 | 479.52 | <.0001 |
| country_c | 5 | 988.43314 | 197.68663 | 5.14 | 0.0002 |
| Source | DF | Type III SS | Mean Square | F Value | Pr > F |
|---|---|---|---|---|---|
| urbanrate_c | 1 | 25.07077 | 25.07077 | 0.65 | 0.4208 |
| incomeperperson_c | 1 | 7.92454 | 7.92454 | 0.21 | 0.6505 |
| internetuserate_c | 1 | 18.45581 | 18.45581 | 0.48 | 0.4896 |
| polityscore_c | 1 | 216.34246 | 216.34246 | 5.63 | 0.0191 |
| lifeexpectancy_c | 1 | 108.05325 | 108.05325 | 2.81 | 0.0959 |
| employrate_c | 1 | 17361.33158 | 17361.33158 | 451.59 | <.0001 |
| country_c | 5 | 988.43314 | 197.68663 | 5.14 | 0.0002 |
| Parameter | Estimate | Standard Error | t Value | Pr > |t| | |
|---|---|---|---|---|---|
| Intercept | 47.41351142 | B | 2.82392755 | 16.79 | <.0001 |
| urbanrate_c | -0.03051835 | 0.03779169 | -0.81 | 0.4208 | |
| incomeperperson_c | -0.00004507 | 0.00009926 | -0.45 | 0.6505 | |
| internetuserate_c | 0.03357976 | 0.04846522 | 0.69 | 0.4896 | |
| polityscore_c | 0.23990981 | 0.10113380 | 2.37 | 0.0191 | |
| lifeexpectancy_c | -0.19790675 | 0.11804861 | -1.68 | 0.0959 | |
| employrate_c | 1.29068372 | 0.06073619 | 21.25 | <.0001 | |
| country_c 0 | -1.90130435 | B | 3.28376933 | -0.58 | 0.5635 |
| country_c 1 | -2.55553870 | B | 3.01657031 | -0.85 | 0.3984 |
| country_c 2 | 6.06544348 | B | 3.12448527 | 1.94 | 0.0543 |
| country_c 3 | -3.45871496 | B | 3.27091579 | -1.06 | 0.2922 |
| country_c 5 | -1.15355514 | B | 3.50164499 | -0.33 | 0.7423 |
| country_c 4 | 0.00000000 | B | . | . | . |
The X'X matrix has been found to be singular, and a generalized inverse was used to solve the normal equations. Terms whose estimates are followed by the letter 'B' are not uniquely estimable.
The GLM Procedure
| Class Level Information | ||
|---|---|---|
| Class | Levels | Values |
| country_c | 6 | 0 1 2 3 4 5 |
| Number of Observations Read | 213 |
|---|---|
| Number of Observations Used | 148 |
The GLM Procedure
Dependent Variable: femaleemployrate
| Source | DF | Sum of Squares | Mean Square | F Value | Pr > F |
|---|---|---|---|---|---|
| Model | 11 | 26988.92492 | 2453.53863 | 63.82 | <.0001 |
| Error | 136 | 5228.50225 | 38.44487 | ||
| Corrected Total | 147 | 32217.42717 |
| R-Square | Coeff Var | Root MSE | femaleemployrate Mean |
|---|---|---|---|
| 0.837712 | 12.89770 | 6.200393 | 48.07365 |
| Source | DF | Type I SS | Mean Square | F Value | Pr > F |
|---|---|---|---|---|---|
| urbanrate_c | 1 | 3885.57081 | 3885.57081 | 101.07 | <.0001 |
| incomeperperson_c | 1 | 1885.43020 | 1885.43020 | 49.04 | <.0001 |
| internetuserate_c | 1 | 8.39997 | 8.39997 | 0.22 | 0.6409 |
| polityscore_c | 1 | 161.21451 | 161.21451 | 4.19 | 0.0425 |
| lifeexpectancy_c | 1 | 1624.64281 | 1624.64281 | 42.26 | <.0001 |
| employrate_c | 1 | 18435.23349 | 18435.23349 | 479.52 | <.0001 |
| country_c | 5 | 988.43314 | 197.68663 | 5.14 | 0.0002 |
| Source | DF | Type III SS | Mean Square | F Value | Pr > F |
|---|---|---|---|---|---|
| urbanrate_c | 1 | 25.07077 | 25.07077 | 0.65 | 0.4208 |
| incomeperperson_c | 1 | 7.92454 | 7.92454 | 0.21 | 0.6505 |
| internetuserate_c | 1 | 18.45581 | 18.45581 | 0.48 | 0.4896 |
| polityscore_c | 1 | 216.34246 | 216.34246 | 5.63 | 0.0191 |
| lifeexpectancy_c | 1 | 108.05325 | 108.05325 | 2.81 | 0.0959 |
| employrate_c | 1 | 17361.33158 | 17361.33158 | 451.59 | <.0001 |
| country_c | 5 | 988.43314 | 197.68663 | 5.14 | 0.0002 |
| Parameter | Estimate | Standard Error | t Value | Pr > |t| | |
|---|---|---|---|---|---|
| Intercept | 46.25995627 | B | 2.22078959 | 20.83 | <.0001 |
| urbanrate_c | -0.03051835 | 0.03779169 | -0.81 | 0.4208 | |
| incomeperperson_c | -0.00004507 | 0.00009926 | -0.45 | 0.6505 | |
| internetuserate_c | 0.03357976 | 0.04846522 | 0.69 | 0.4896 | |
| polityscore_c | 0.23990981 | 0.10113380 | 2.37 | 0.0191 | |
| lifeexpectancy_c | -0.19790675 | 0.11804861 | -1.68 | 0.0959 | |
| employrate_c | 1.29068372 | 0.06073619 | 21.25 | <.0001 | |
| country_c 0 | -0.74774921 | B | 2.82427232 | -0.26 | 0.7916 |
| country_c 1 | -1.40198356 | B | 2.46450008 | -0.57 | 0.5704 |
| country_c 2 | 7.21899862 | B | 2.59052312 | 2.79 | 0.0061 |
| country_c 3 | -2.30515982 | B | 2.62103664 | -0.88 | 0.3807 |
| country_c 4 | 1.15355514 | B | 3.50164499 | 0.33 | 0.7423 |
| country_c 5 | 0.00000000 | B | . | . | . |
The X'X matrix has been found to be singular, and a generalized inverse was used to solve the normal equations. Terms whose estimates are followed by the letter 'B' are not uniquely estimable.
The LOGISTIC Procedure
| Model Information | |
|---|---|
| Data Set | WORK.NEW2 |
| Response Variable | fememploy |
| Number of Response Levels | 2 |
| Model | binary logit |
| Optimization Technique | Fisher's scoring |
| Number of Observations Read | 213 |
|---|---|
| Number of Observations Used | 203 |
| Response Profile | ||
|---|---|---|
| Ordered Value | fememploy | Total Frequency |
| 1 | 1 | 76 |
| 2 | 0 | 127 |
Probability modeled is fememploy=1.
10 observations were deleted due to missing values for the response or explanatory variables.
| Model Convergence Status |
|---|
| Convergence criterion (GCONV=1E-8) satisfied. |
| Model Fit Statistics | ||
|---|---|---|
| Criterion | Intercept Only | Intercept and Covariates |
| AIC | 270.467 | 265.476 |
| SC | 273.780 | 272.102 |
| -2 Log L | 268.467 | 261.476 |
| Testing Global Null Hypothesis: BETA=0 | |||
|---|---|---|---|
| Test | Chi-Square | DF | Pr > ChiSq |
| Likelihood Ratio | 6.9908 | 1 | 0.0082 |
| Score | 6.9096 | 1 | 0.0086 |
| Wald | 6.7389 | 1 | 0.0094 |
| Analysis of Maximum Likelihood Estimates | |||||
|---|---|---|---|---|---|
| Parameter | DF | Estimate | Standard Error | Wald Chi-Square | Pr > ChiSq |
| Intercept | 1 | -0.5322 | 0.1483 | 12.8872 | 0.0003 |
| urbanrate_c | 1 | -0.0164 | 0.00632 | 6.7389 | 0.0094 |
| Odds Ratio Estimates | |||
|---|---|---|---|
| Effect | Point Estimate | 95% Wald Confidence Limits | |
| urbanrate_c | 0.984 | 0.972 | 0.996 |
| Association of Predicted Probabilities and Observed Responses | |||
|---|---|---|---|
| Percent Concordant | 60.5 | Somers' D | 0.211 |
| Percent Discordant | 39.4 | Gamma | 0.211 |
| Percent Tied | 0.1 | Tau-a | 0.099 |
| Pairs | 9652 | c | 0.605 |
The LOGISTIC Procedure
| Model Information | |
|---|---|
| Data Set | WORK.NEW2 |
| Response Variable | fememploy |
| Number of Response Levels | 2 |
| Model | binary logit |
| Optimization Technique | Fisher's scoring |
| Number of Observations Read | 213 |
|---|---|
| Number of Observations Used | 189 |
| Response Profile | ||
|---|---|---|
| Ordered Value | fememploy | Total Frequency |
| 1 | 1 | 73 |
| 2 | 0 | 116 |
Probability modeled is fememploy=1.
24 observations were deleted due to missing values for the response or explanatory variables.
| Model Convergence Status |
|---|
| Convergence criterion (GCONV=1E-8) satisfied. |
| Model Fit Statistics | ||
|---|---|---|
| Criterion | Intercept Only | Intercept and Covariates |
| AIC | 254.140 | 251.476 |
| SC | 257.382 | 261.201 |
| -2 Log L | 252.140 | 245.476 |
| Testing Global Null Hypothesis: BETA=0 | |||
|---|---|---|---|
| Test | Chi-Square | DF | Pr > ChiSq |
| Likelihood Ratio | 6.6645 | 2 | 0.0357 |
| Score | 6.6145 | 2 | 0.0366 |
| Wald | 6.3006 | 2 | 0.0428 |
| Analysis of Maximum Likelihood Estimates | |||||
|---|---|---|---|---|---|
| Parameter | DF | Estimate | Standard Error | Wald Chi-Square | Pr > ChiSq |
| Intercept | 1 | -0.4864 | 0.1527 | 10.1406 | 0.0015 |
| urbanrate_c | 1 | -0.0191 | 0.00762 | 6.2733 | 0.0123 |
| incomeperperson_c | 1 | 0.000015 | 0.000013 | 1.3882 | 0.2387 |
| Odds Ratio Estimates | |||
|---|---|---|---|
| Effect | Point Estimate | 95% Wald Confidence Limits | |
| urbanrate_c | 0.981 | 0.967 | 0.996 |
| incomeperperson_c | 1.000 | 1.000 | 1.000 |
| Association of Predicted Probabilities and Observed Responses | |||
|---|---|---|---|
| Percent Concordant | 61.8 | Somers' D | 0.235 |
| Percent Discordant | 38.2 | Gamma | 0.235 |
| Percent Tied | 0.0 | Tau-a | 0.112 |
| Pairs | 8468 | c | 0.618 |
The LOGISTIC Procedure
| Model Information | |
|---|---|
| Data Set | WORK.NEW2 |
| Response Variable | fememploy |
| Number of Response Levels | 2 |
| Model | binary logit |
| Optimization Technique | Fisher's scoring |
| Number of Observations Read | 213 |
|---|---|
| Number of Observations Used | 182 |
| Response Profile | ||
|---|---|---|
| Ordered Value | fememploy | Total Frequency |
| 1 | 1 | 72 |
| 2 | 0 | 110 |
Probability modeled is fememploy=1.
31 observations were deleted due to missing values for the response or explanatory variables.
| Model Convergence Status |
|---|
| Convergence criterion (GCONV=1E-8) satisfied. |
| Model Fit Statistics | ||
|---|---|---|
| Criterion | Intercept Only | Intercept and Covariates |
| AIC | 246.313 | 244.655 |
| SC | 249.517 | 257.471 |
| -2 Log L | 244.313 | 236.655 |
| Testing Global Null Hypothesis: BETA=0 | |||
|---|---|---|---|
| Test | Chi-Square | DF | Pr > ChiSq |
| Likelihood Ratio | 7.6575 | 3 | 0.0536 |
| Score | 7.4025 | 3 | 0.0601 |
| Wald | 6.7167 | 3 | 0.0815 |
| Analysis of Maximum Likelihood Estimates | |||||
|---|---|---|---|---|---|
| Parameter | DF | Estimate | Standard Error | Wald Chi-Square | Pr > ChiSq |
| Intercept | 1 | -0.4352 | 0.1555 | 7.8326 | 0.0051 |
| urbanrate_c | 1 | -0.0166 | 0.00879 | 3.5531 | 0.0594 |
| incomeperperson_c | 1 | 0.000038 | 0.000021 | 3.1057 | 0.0780 |
| internetuserate_c | 1 | -0.00886 | 0.00958 | 0.8564 | 0.3548 |
| Odds Ratio Estimates | |||
|---|---|---|---|
| Effect | Point Estimate | 95% Wald Confidence Limits | |
| urbanrate_c | 0.984 | 0.967 | 1.001 |
| incomeperperson_c | 1.000 | 1.000 | 1.000 |
| internetuserate_c | 0.991 | 0.973 | 1.010 |
| Association of Predicted Probabilities and Observed Responses | |||
|---|---|---|---|
| Percent Concordant | 65.5 | Somers' D | 0.310 |
| Percent Discordant | 34.5 | Gamma | 0.310 |
| Percent Tied | 0.0 | Tau-a | 0.149 |
| Pairs | 7920 | c | 0.655 |
The LOGISTIC Procedure
| Model Information | |
|---|---|
| Data Set | WORK.NEW2 |
| Response Variable | fememploy |
| Number of Response Levels | 2 |
| Model | binary logit |
| Optimization Technique | Fisher's scoring |
| Number of Observations Read | 213 |
|---|---|
| Number of Observations Used | 153 |
| Response Profile | ||
|---|---|---|
| Ordered Value | fememploy | Total Frequency |
| 1 | 1 | 66 |
| 2 | 0 | 87 |
Probability modeled is fememploy=1.
60 observations were deleted due to missing values for the response or explanatory variables.
| Model Convergence Status |
|---|
| Convergence criterion (GCONV=1E-8) satisfied. |
| Model Fit Statistics | ||
|---|---|---|
| Criterion | Intercept Only | Intercept and Covariates |
| AIC | 211.212 | 198.007 |
| SC | 214.242 | 213.159 |
| -2 Log L | 209.212 | 188.007 |
| Testing Global Null Hypothesis: BETA=0 | |||
|---|---|---|---|
| Test | Chi-Square | DF | Pr > ChiSq |
| Likelihood Ratio | 21.2043 | 4 | 0.0003 |
| Score | 20.2485 | 4 | 0.0004 |
| Wald | 18.2513 | 4 | 0.0011 |
| Analysis of Maximum Likelihood Estimates | |||||
|---|---|---|---|---|---|
| Parameter | DF | Estimate | Standard Error | Wald Chi-Square | Pr > ChiSq |
| Intercept | 1 | -0.2135 | 0.1799 | 1.4086 | 0.2353 |
| urbanrate_c | 1 | -0.0380 | 0.0113 | 11.3202 | 0.0008 |
| incomeperperson_c | 1 | 0.000093 | 0.000032 | 8.2393 | 0.0041 |
| internetuserate_c | 1 | -0.0121 | 0.0127 | 0.9050 | 0.3414 |
| polityscore_c | 1 | -0.00425 | 0.0302 | 0.0198 | 0.8880 |
| Odds Ratio Estimates | |||
|---|---|---|---|
| Effect | Point Estimate | 95% Wald Confidence Limits | |
| urbanrate_c | 0.963 | 0.942 | 0.984 |
| incomeperperson_c | 1.000 | 1.000 | 1.000 |
| internetuserate_c | 0.988 | 0.964 | 1.013 |
| polityscore_c | 0.996 | 0.939 | 1.056 |
| Association of Predicted Probabilities and Observed Responses | |||
|---|---|---|---|
| Percent Concordant | 71.5 | Somers' D | 0.429 |
| Percent Discordant | 28.5 | Gamma | 0.429 |
| Percent Tied | 0.0 | Tau-a | 0.212 |
| Pairs | 5742 | c | 0.715 |
The LOGISTIC Procedure
| Model Information | |
|---|---|
| Data Set | WORK.NEW2 |
| Response Variable | fememploy |
| Number of Response Levels | 2 |
| Model | binary logit |
| Optimization Technique | Fisher's scoring |
| Number of Observations Read | 213 |
|---|---|
| Number of Observations Used | 153 |
| Response Profile | ||
|---|---|---|
| Ordered Value | fememploy | Total Frequency |
| 1 | 1 | 66 |
| 2 | 0 | 87 |
Probability modeled is fememploy=1.
60 observations were deleted due to missing values for the response or explanatory variables.
| Model Convergence Status |
|---|
| Convergence criterion (GCONV=1E-8) satisfied. |
| Model Fit Statistics | ||
|---|---|---|
| Criterion | Intercept Only | Intercept and Covariates |
| AIC | 211.212 | 198.007 |
| SC | 214.242 | 213.159 |
| -2 Log L | 209.212 | 188.007 |
| Testing Global Null Hypothesis: BETA=0 | |||
|---|---|---|---|
| Test | Chi-Square | DF | Pr > ChiSq |
| Likelihood Ratio | 21.2043 | 4 | 0.0003 |
| Score | 20.2485 | 4 | 0.0004 |
| Wald | 18.2513 | 4 | 0.0011 |
| Analysis of Maximum Likelihood Estimates | |||||
|---|---|---|---|---|---|
| Parameter | DF | Estimate | Standard Error | Wald Chi-Square | Pr > ChiSq |
| Intercept | 1 | -0.2135 | 0.1799 | 1.4086 | 0.2353 |
| urbanrate_c | 1 | -0.0380 | 0.0113 | 11.3202 | 0.0008 |
| incomeperperson_c | 1 | 0.000093 | 0.000032 | 8.2393 | 0.0041 |
| internetuserate_c | 1 | -0.0121 | 0.0127 | 0.9050 | 0.3414 |
| polityscore_c | 1 | -0.00425 | 0.0302 | 0.0198 | 0.8880 |
| Odds Ratio Estimates | |||
|---|---|---|---|
| Effect | Point Estimate | 95% Wald Confidence Limits | |
| urbanrate_c | 0.963 | 0.942 | 0.984 |
| incomeperperson_c | 1.000 | 1.000 | 1.000 |
| internetuserate_c | 0.988 | 0.964 | 1.013 |
| polityscore_c | 0.996 | 0.939 | 1.056 |
| Association of Predicted Probabilities and Observed Responses | |||
|---|---|---|---|
| Percent Concordant | 71.5 | Somers' D | 0.429 |
| Percent Discordant | 28.5 | Gamma | 0.429 |
| Percent Tied | 0.0 | Tau-a | 0.212 |
| Pairs | 5742 | c | 0.715 |
The LOGISTIC Procedure
| Model Information | |
|---|---|
| Data Set | WORK.NEW2 |
| Response Variable | fememploy |
| Number of Response Levels | 2 |
| Model | binary logit |
| Optimization Technique | Fisher's scoring |
| Number of Observations Read | 213 |
|---|---|
| Number of Observations Used | 153 |
| Response Profile | ||
|---|---|---|
| Ordered Value | fememploy | Total Frequency |
| 1 | 1 | 66 |
| 2 | 0 | 87 |
Probability modeled is fememploy=1.
60 observations were deleted due to missing values for the response or explanatory variables.
| Model Convergence Status |
|---|
| Convergence criterion (GCONV=1E-8) satisfied. |
| Model Fit Statistics | ||
|---|---|---|
| Criterion | Intercept Only | Intercept and Covariates |
| AIC | 211.212 | 196.396 |
| SC | 214.242 | 214.579 |
| -2 Log L | 209.212 | 184.396 |
| Testing Global Null Hypothesis: BETA=0 | |||
|---|---|---|---|
| Test | Chi-Square | DF | Pr > ChiSq |
| Likelihood Ratio | 24.8153 | 5 | 0.0002 |
| Score | 23.6435 | 5 | 0.0003 |
| Wald | 21.0038 | 5 | 0.0008 |
| Analysis of Maximum Likelihood Estimates | |||||
|---|---|---|---|---|---|
| Parameter | DF | Estimate | Standard Error | Wald Chi-Square | Pr > ChiSq |
| Intercept | 1 | -0.2208 | 0.1822 | 1.4684 | 0.2256 |
| urbanrate_c | 1 | -0.0333 | 0.0116 | 8.2216 | 0.0041 |
| incomeperperson_c | 1 | 0.000088 | 0.000032 | 7.6019 | 0.0058 |
| internetuserate_c | 1 | 0.00154 | 0.0146 | 0.0111 | 0.9159 |
| polityscore_c | 1 | -0.00040 | 0.0309 | 0.0002 | 0.9897 |
| lifeexpectancy_c | 1 | -0.0549 | 0.0294 | 3.4985 | 0.0614 |
| Odds Ratio Estimates | |||
|---|---|---|---|
| Effect | Point Estimate | 95% Wald Confidence Limits | |
| urbanrate_c | 0.967 | 0.946 | 0.990 |
| incomeperperson_c | 1.000 | 1.000 | 1.000 |
| internetuserate_c | 1.002 | 0.973 | 1.031 |
| polityscore_c | 1.000 | 0.941 | 1.062 |
| lifeexpectancy_c | 0.947 | 0.894 | 1.003 |
| Association of Predicted Probabilities and Observed Responses | |||
|---|---|---|---|
| Percent Concordant | 73.0 | Somers' D | 0.459 |
| Percent Discordant | 27.0 | Gamma | 0.459 |
| Percent Tied | 0.0 | Tau-a | 0.227 |
| Pairs | 5742 | c | 0.730 |
The LOGISTIC Procedure
| Model Information | |
|---|---|
| Data Set | WORK.NEW2 |
| Response Variable | fememploy |
| Number of Response Levels | 2 |
| Model | binary logit |
| Optimization Technique | Fisher's scoring |
| Number of Observations Read | 213 |
|---|---|
| Number of Observations Used | 150 |
| Response Profile | ||
|---|---|---|
| Ordered Value | fememploy | Total Frequency |
| 1 | 1 | 66 |
| 2 | 0 | 84 |
Probability modeled is fememploy=1.
63 observations were deleted due to missing values for the response or explanatory variables.
| Model Convergence Status |
|---|
| Convergence criterion (GCONV=1E-8) satisfied. |
| Model Fit Statistics | ||
|---|---|---|
| Criterion | Intercept Only | Intercept and Covariates |
| AIC | 207.779 | 122.260 |
| SC | 210.790 | 143.334 |
| -2 Log L | 205.779 | 108.260 |
| Testing Global Null Hypothesis: BETA=0 | |||
|---|---|---|---|
| Test | Chi-Square | DF | Pr > ChiSq |
| Likelihood Ratio | 97.5193 | 6 | <.0001 |
| Score | 69.8546 | 6 | <.0001 |
| Wald | 33.8788 | 6 | <.0001 |
| Analysis of Maximum Likelihood Estimates | |||||
|---|---|---|---|---|---|
| Parameter | DF | Estimate | Standard Error | Wald Chi-Square | Pr > ChiSq |
| Intercept | 1 | -0.6330 | 0.2649 | 5.7095 | 0.0169 |
| urbanrate_c | 1 | -0.0171 | 0.0156 | 1.1935 | 0.2746 |
| incomeperperson_c | 1 | 4.544E-6 | 0.000039 | 0.0134 | 0.9077 |
| internetuserate_c | 1 | 0.0213 | 0.0191 | 1.2462 | 0.2643 |
| polityscore_c | 1 | 0.0488 | 0.0485 | 1.0117 | 0.3145 |
| lifeexpectancy_c | 1 | -0.0223 | 0.0437 | 0.2607 | 0.6097 |
| employrate_c | 1 | 0.2924 | 0.0543 | 28.9439 | <.0001 |
| Odds Ratio Estimates | |||
|---|---|---|---|
| Effect | Point Estimate | 95% Wald Confidence Limits | |
| urbanrate_c | 0.983 | 0.953 | 1.014 |
| incomeperperson_c | 1.000 | 1.000 | 1.000 |
| internetuserate_c | 1.022 | 0.984 | 1.060 |
| polityscore_c | 1.050 | 0.955 | 1.155 |
| lifeexpectancy_c | 0.978 | 0.898 | 1.065 |
| employrate_c | 1.340 | 1.204 | 1.490 |
| Association of Predicted Probabilities and Observed Responses | |||
|---|---|---|---|
| Percent Concordant | 92.4 | Somers' D | 0.848 |
| Percent Discordant | 7.6 | Gamma | 0.848 |
| Percent Tied | 0.0 | Tau-a | 0.421 |
| Pairs | 5544 | c | 0.924 |
