Testing a Potential Moderator
In this case, I am using the ANOVA tool for assessing the following hypothesis:
"There is a relationship between drink beer and be laid off at the last 12 months, as moderator is the situation to have got separated or broken off a relationship"
Quantity response variable=Quantity beer
Categorical explanatory variable=Be laid off or not
Moderator=got separated or broken off a relationship
In the first case when a person has got separated or broken off a relationship, p value =0.0735, that is it is not statistically significantly, for the second case p-value=.0001 it reveals that there is a relationship between the variables.
In general in the media values reveal for both cases either with or without to have got separated or broken relationship, people who has been laid off tend to drink more than who has a job. This can see in the group S1Q234(Laid off 1 yes, 2 no) level 1 is bigger than level 2 in both cases
"There is a relationship between drink beer and be laid off at the last 12 months, as moderator is the situation to have got separated or broken off a relationship"
Quantity response variable=Quantity beer
Categorical explanatory variable=Be laid off or not
Moderator=got separated or broken off a relationship
In the first case when a person has got separated or broken off a relationship, p value =0.0735, that is it is not statistically significantly, for the second case p-value=.0001 it reveals that there is a relationship between the variables.
In general in the media values reveal for both cases either with or without to have got separated or broken relationship, people who has been laid off tend to drink more than who has a job. This can see in the group S1Q234(Laid off 1 yes, 2 no) level 1 is bigger than level 2 in both cases
LIBNAME mydata "/courses/d1406ae5ba27fe300 " access=readonly;
DATA new; set mydata.nesarc_pds;
LABEL ALCABDEP12DX="Alcohol Dependence Past 12 Months"
S2AQ5A=" Drank any beer Last 12 Months"
S2AQ5B="How often Drank beer Last 12 Months"
S2AQ5D="Numbers of beer consumed";
/*Set appropriate missing data as needed*/
IF S2AQ5B=99 THEN S2AQ5B=.;
IF S2AQ5D=99 THEN S2AQ5D=.;
IF S2AQ5B=1 THEN FREQMO=365;
ELSE IF S2AQ5B=2 THEN FREQMO=312;
ELSE IF S2AQ5B=3 THEN FREQMO=208;
ELSE IF S2AQ5B=4 THEN FREQMO=104;
ELSE IF S2AQ5B=5 THEN FREQMO=52;
ELSE IF S2AQ5B=6 THEN FREQMO=36;
ELSE IF S2AQ5B=7 THEN FREQMO=12;
ELSE IF S2AQ5B=8 THEN FREQMO=11;
ELSE IF S2AQ5B=9 THEN FREQMO=6;
ELSE IF S2AQ5B=10 THEN FREQMO=2;
/*days per year
365=Every day
312=Nearly every day
208=3 to 4 times a week
104=2 times a week
52=Once a week
36=2 to 3 times a month
12=Once a month
11=7 to 11 times in the last year
6=3 to 6 times in the last year
2=1 or 2 times in the last year*/
NUMBEER=FREQMO*S2AQ5D;
/*subsetting data to include only past 12 month smokers, age 18-25*/
IF S2AQ5A=1;
IF AGE LE 25;
PROC SORT; by IDNUM;
PROC SORT; by S1Q238;
PROC ANOVA; CLASS S1Q234;
MODEL NUMBEER=S1Q234;
MEANS S1Q234;by S1Q238;
RUN;
DATA new; set mydata.nesarc_pds;
LABEL ALCABDEP12DX="Alcohol Dependence Past 12 Months"
S2AQ5A=" Drank any beer Last 12 Months"
S2AQ5B="How often Drank beer Last 12 Months"
S2AQ5D="Numbers of beer consumed";
/*Set appropriate missing data as needed*/
IF S2AQ5B=99 THEN S2AQ5B=.;
IF S2AQ5D=99 THEN S2AQ5D=.;
IF S2AQ5B=1 THEN FREQMO=365;
ELSE IF S2AQ5B=2 THEN FREQMO=312;
ELSE IF S2AQ5B=3 THEN FREQMO=208;
ELSE IF S2AQ5B=4 THEN FREQMO=104;
ELSE IF S2AQ5B=5 THEN FREQMO=52;
ELSE IF S2AQ5B=6 THEN FREQMO=36;
ELSE IF S2AQ5B=7 THEN FREQMO=12;
ELSE IF S2AQ5B=8 THEN FREQMO=11;
ELSE IF S2AQ5B=9 THEN FREQMO=6;
ELSE IF S2AQ5B=10 THEN FREQMO=2;
/*days per year
365=Every day
312=Nearly every day
208=3 to 4 times a week
104=2 times a week
52=Once a week
36=2 to 3 times a month
12=Once a month
11=7 to 11 times in the last year
6=3 to 6 times in the last year
2=1 or 2 times in the last year*/
NUMBEER=FREQMO*S2AQ5D;
/*subsetting data to include only past 12 month smokers, age 18-25*/
IF S2AQ5A=1;
IF AGE LE 25;
PROC SORT; by IDNUM;
PROC SORT; by S1Q238;
PROC ANOVA; CLASS S1Q234;
MODEL NUMBEER=S1Q234;
MEANS S1Q234;by S1Q238;
RUN;
Procedimiento ANOVA
| Información de nivel de clase | ||
|---|---|---|
| Clase | Niveles | Valores |
| S1Q234 | 3 | 1 2 9 |
| Número de observaciones leídas | 494 |
|---|---|
| Número de observaciones usadas | 491 |
Procedimiento ANOVA
Variable dependiente: NUMBEER
| Fuente | DF | Suma de cuadrados | Cuadrado de la media | F-Valor | Pr > F |
|---|---|---|---|---|---|
| Modelo | 2 | 3969215.1 | 1984607.6 | 2.62 | 0.0735 |
| Error | 488 | 369085652.9 | 756323.1 | ||
| Total corregido | 490 | 373054868.0 |
| R-cuadrado | Coef Var | Raíz MSE | NUMBEER Media |
|---|---|---|---|
| 0.010640 | 201.4764 | 869.6684 | 431.6477 |
| Fuente | DF | Anova SS | Cuadrado de la media | F-Valor | Pr > F |
|---|---|---|---|---|---|
| S1Q234 | 2 | 3969215.143 | 1984607.572 | 2.62 | 0.0735 |
Procedimiento ANOVA
| Nivel de S1Q234 |
N | NUMBEER | |
|---|---|---|---|
| Media | Dev std | ||
| 1 | 102 | 605.029412 | 1215.06217 |
| 2 | 388 | 386.994845 | 753.92456 |
| 9 | 1 | 72.000000 | . |
Procedimiento ANOVA
| Información de nivel de clase | ||
|---|---|---|
| Clase | Niveles | Valores |
| S1Q234 | 2 | 1 2 |
| Número de observaciones leídas | 2357 |
|---|---|
| Número de observaciones usadas | 2348 |
Procedimiento ANOVA
Variable dependiente: NUMBEER
| Fuente | DF | Suma de cuadrados | Cuadrado de la media | F-Valor | Pr > F |
|---|---|---|---|---|---|
| Modelo | 1 | 6406134.2 | 6406134.2 | 15.21 | <.0001 |
| Error | 2346 | 988331490.7 | 421283.7 | ||
| Total corregido | 2347 | 994737624.9 |
| R-cuadrado | Coef Var | Raíz MSE | NUMBEER Media |
|---|---|---|---|
| 0.006440 | 216.3500 | 649.0637 | 300.0064 |
| Fuente | DF | Anova SS | Cuadrado de la media | F-Valor | Pr > F |
|---|---|---|---|---|---|
| S1Q234 | 1 | 6406134.184 | 6406134.184 | 15.21 | <.0001 |
Procedimiento ANOVA
| Nivel de S1Q234 |
N | NUMBEER | |
|---|---|---|---|
| Media | Dev std | ||
| 1 | 293 | 438.337884 |
|---|
| 1079.88294 |
| 2 | 2055 | 280.283212 |
|---|
| 561.59831 |
