In: Statistics and Probability
To understand the implications of expanding health coverage, researchers want to know if people with health insurance use more health care than do people without it. To test this, they leveraged an administrative lottery that randomly assigned individuals to either receive public health care or not. Using the data below on the number of medical interactions (e.g., hospital visits or doctor appointments), test whether individuals with health insurance use health services differently than do those without it (alpha=0.1, two tails).
No Insurance | Insurance |
2 | 3 |
1 | 4 |
0 | 3 |
3 | 3 |
2 | 2 |
LET BE THE MEAN FOR INDIVIDUALS WITHOUT INSURANCE AND BE THE MEAN FOR INDIVIDUALS WITH INSURANCE. WE WANT TO TEST IF INDIVIDUALS WITH HEALTH INSURANCE USE HEALTH SERVICE DIFFERENTLY THAN WITHOUT IT.
SO THE HYPOTHESIS HERE IS,
WE USE A TWO SAMPLE T-TEST AT LEVEL OF SIGNIFICANCE 0.1 AS THERE IS INDEPENDENCE IN THE DATA.WE USE MINITAB 16 TO CARRY OUT THE HYPOTHESIS TESTING,
STEPS- ENTER THE DATA IN SEPERATE COLUMNS> STAT> BASIC STATISTICS> TWO SAMPLE-T> SELECT THE SAMPLES> UNDER 'OPTIONS', WE SET THE CONFIDENCE LEVEL AS 90.0 AND ALTERNATE AS 'NOT EQUAL'> OK.
OBSERVATION-
THE TEST STATISTIC OBTAINED IS T= -2.33
THE CORRESPONDING P-VALUE IS = 0.058
AS THE P-VALUE< LEVEL OF SIGNIFICANCE, WE REJECT THE NULL HYPOTHESIS AND CONCLUDE THAT THE DATA SETS ARE SIGNIFICANTLY AT