I have the rate of vaccination in my county for a six week period before an intervention took place and want to compare it to a six week period after the intervention took place. I am looking to determine if there was a significant increase in the vaccination rate due to the intervention. I thought this would be a paired sample t test since we are measuring the same population before and after an intervention but my professor is saying it has to be an independent t test because I am not measuring the same people twice. When I look this up, it says the independent t test is done when you have a control group and a treatment group. That is not what I have and I am confused. How do you think this should be set up?
Which test to use? Measuring rate of vaccination on a population before and after intervention.
- Thread starter o2bdone
- Start date
How will you determine that any changes over time are due to the intervention? Which additional
information wil you include which makes causal claims defendable?
You have data for a whole country, why would you want to perform inferential
statistics, i.e. statistics which make inferences from samples to populations?
Apart from this, you dependent variable is categorical (vaccinated yes/no).
t-Tests are for interval scaled variables such as height, weight, IQ etc.
Could you please tell us why you have to work on this problem? Is it homework for
school or something like that?
With kind regards
Karabiner
information wil you include which makes causal claims defendable?
You have data for a whole country, why would you want to perform inferential
statistics, i.e. statistics which make inferences from samples to populations?
Apart from this, you dependent variable is categorical (vaccinated yes/no).
t-Tests are for interval scaled variables such as height, weight, IQ etc.
Could you please tell us why you have to work on this problem? Is it homework for
school or something like that?
With kind regards
Karabiner
Thank you for responding! I am a student in college and as part of my internship, I was asked to assist a county health department in increasing their vaccination rate against Hep A. Their goal was to reach 80% of the target population within their county. At that time, their rate of vaccination rate was only 15%. I located partners in town to join with in the effort and provided posters, flyers, and resources needed. Following the intervention, we began to see an increase in the rate. 6 months later, we reached the desired 80% of the county. I am now trying to write this up and have went back and forth with my professor on the test and set up. I am beginning to feel as though we are just not communicating well so I thought I would seek another opinion. What I have are weekly reports showing the counts of vaccinations done that week as well as a weekly report from the state office showing the rate for our county that week. No identifying information was used for each patient, just a number of vaccinations per facility per week. For example, our county jail became a partner. Prior to our intervention, the County Health Department was the only county partner participating in the effort. After signing them up to assist, training them, providing them with resources needed, they began vaccinating along with us and the rate gradually increased. We added 4 partners in total. I have listed all counts in excel spreadsheets for every week during a six month period along with weekly reports showing the percent of the target population that had been vaccinated. I thought I could do a paired samples t-test on the county rate. My professor explained that you can only due this when measuring the same people twice. In my mind, these are the same people because the people are "my county target population" and it is being measured twice, before and after an intervention. He is saying it must be an independent t- test to show the difference between the control and treatment group. I don't see the control group. I feel there is a better way to set this up. I am not measuring individuals, I am measuring the total vaccination rate at two different points. Thank you for your help!