what statistical method do I need to use?

#1
Hello there sir/ma'am. I am presently conducting a research study which uses statistical procedure. I already gathered data using survey instrument. Unfortunately, I didn't finish gettting samples due to quarantine. anyone may help what possible to do? do I still need to use slovins for my sample size? or clustering? 15 schools which one classroom per school was surveyed. any help is appreciated. thanks.
 

Karabiner

TS Contributor
#2
You maybe could describe your study in more detail (topic, research questions, study design, variables, measurement instruments, statistical tests planned, sample size achieved).

With kind regards

Karabiner
 
#3
Hello Karabiner, thank you for the response.

the topic of my study is about on the popularity awareness of Geodetic Engineering course in Pampanga. I got 396 sample size by using the Slovin's formula. I used Quantitative type of research by using Likert scale, yes-no question and by means of ranking. I'm planning to use Chronbach's alpha for testing the reliability of the gathered data friedman test for the ranking and for the yes-no question I'll use percentage pie chart.
 
#4
I used multi stage sampling for my procedure. I cluster the 22 Municipalities into four groups. 5-6-5-6 by using systematic random sampling. Next was I used Stratified random sampling to get the respected samples needed in each cluster from the total samples I got by using the Slovin's Formula. I have a population of 38126 students from 133 different schools.

Best regards

Joseph
 

Karabiner

TS Contributor
#5
I didn't know that Slovin's formula could be applied for clustered data. Anyway, which alternatives do you have?
Wait until the epidemic is over and complete the study afterwards? If not, you have to coontact your supervisor
if it's ok to do the analyses with a smaller sample size. You could do a power analysis concerning your main
outcome variable. In that power analysis you can plug in your achieved sample size and then will calculate the power
to detect a small or large effect with that sample size.

With kind regards

Karabiner
 
#6
Hi, sorry for that. I just taught it would be possible if i'll use slovin's for clustered data. Anyways, thank you for the suggestion. I have searched about the power analysis and still reading about it. May i know what power analysis could be used? such what formula to be used? thanks for the help.

Best regards,

Joseph