Statstics Assignmet
Use Excel, Megstat, StatCrunch, or any statistical tool you would like to answer the questions presented.
Suppose we worked for a provider of Urgent Care Centers, and we wanted to control the costs of medication inventory. One way to do that is to stock medication in proportion to its usage in the clinics. As part of this study, we wanted to compare the proportions of different NSAID's (Non-Steroid Anti-Inflammatory Drugs) dispensed to patients to see if one is more frequently dispensed than others. So, the categories may be Aspirin, Acetaminophen, Ibuprofen, Naproxen, and Others. Knowing this information would help us manage inventory costs better by ordering NSAID's in proportion to their issuance at the clinics.
So:
RQ: Do doctors dispense one type of NSAID more than others?
HO: Doctors show no preference in the NSAID they dispense to patients.
HA: Doctors show a preference in the NSAID they dispense to patients.
If there is no difference in preference, we would expect that all 5 categories would be dispensed in the same proportion over a specific timeframe to affected patients. So the expected distribution would be:
• 20% Aspirin
• 20% Acetaminophen
• 20% Ibuprofen
• 20% Naproxen
• 20% Others
Let's say after 30 days, 100 patients were given NSAIDS as part of their treatment, and the observed distribution was as follows:
• 5 Aspirin
• 10 Acetaminophen
• 30 Ibuprofen
• 50 Naproxen
• 5 Others
We would then set up the chi-square analysis to test for a statistically significant difference between the observed distribution and the expected distribution.
To the class:
• What would the expected frequency distribution look like for the 100 observations?
• What would the results of the chi-square analysis be?