Assignment 1: Hypothesis testing and ANOVA

Coursera: Hypothesis testing and ANOVA assignment

1.- Analysis question:

I want to verify whether the topic that states that married people are fatter than single ones is true or not.

NOTE: as I have no previous knowledge of SAS, I stuck to a question that allowed me to complete the assignment by using as less code as possible. Sorry for the simplicity of the exercise.

I have used the NESARC data set to compare the WEIGHT field means by the SPOUSE groups.

2.- Code used:

LIBNAME mydata "/courses/d1406ae5ba27fe300 " access=readonly;

DATA new; set mydata.nesarc_pds;

LABEL SPOUSE="MARRIED OR UNMARRIED"

                  WEIGHT="WEIGHT IN POUNDS"

PROC SORT; by IDNUM;

PROC ANOVA; CLASS SPOUSE;

MODEL WEIGHT=SPOUSE;

MEANS SPOUSE;

RUN;

3.- Results

Información de nivel de clase
Clase	Niveles	Valores
SPOUSE	2	1 2

Número de observaciones leídas	43093
Número de observaciones usadas	43093

Fuente	DF	Suma de cuadrados	Cuadrado de la media	F-Valor	Pr > F
Modelo	1	51114964804	51114964804	4663.93	<.0001
Error	43091	472261177300	10959624.453
Total corregido	43092	523376142103

R-cuadrado	Coef Var	Raíz MSE	WEIGHT Media
0.097664	68.62655	3310.532	4823.982

Fuente	DF	Anova SS	Cuadrado de la media	F-Valor	Pr > F
SPOUSE	1	51114964804	51114964804	4663.93	<.0001

Nivel de SPOUSE	N	WEIGHT
		Mean	Dev std
1	19810	6004.70424	3536.74459
2	23283	3819.38097	3105.11566

4.- Interpretation:

As the means for both groups are different enough (6004,7 vs 3819.4) and the level of significance is under 0,0001, we can conclude that married people are statistically likely to be fatter than unmarried ones.