Any statistical wizards out there?
I must admit that I am confused on the capability of the Paired T test. Please let me know if I am using it correctly, question is a little detailed but I want to make sure that I understand it:
Say you are you have a group of teachers and a group of students.
The Teachers instructs during the week to traditional high school students and on the weekend to GED students---The subjects are exactly the same that the Teachers cover to the two groups yet the teacher instruct different courses, so the curriculum is exactly the same.
The teachers at the end of the year give their students an opportunity to evaluate the students experience regarding the lessons—giving the student let’s say the ability to rate the difficulty level of the subject matter, content, and assignment load.
The group of teachers wants to see if the difficulty level of the curriculum is equal between the two groups. So they add up the scores of the evaluations. Adding all the scores of the evaluation together to get the scores below (so you have a group of teachers who teach different topics then there fellow teachers but the topics they teach are mirrored in High School and the GED class’s (i.e. the history teacher teaches the same history class to the GED students and the High School students, math teacher and so on) For they are not worried about the class’s as individuals but rather the comparison of GED class’s vs. High School class’s. The goal would be to find if the students in the 2 different groups view the material the same way, given the fact that they are 2 different groups but are being taught the same material by the same teacher.
An example of the math
Subject matter
GED= 10.4 High School=11.1
Content
GED=11 High School=10.5
Assignment Load
GED=13.1 High School=12.8
Group 1 (GED)
10.4, 11, 13.1
Group 2 (HS)
11.1, 10.5, 12.8
If you ran the data you would get:
Paired t test results
P value and statistical significance:
The two-tailed P value equals 0.9366
By conventional criteria; this difference is considered to be not statistically significant.
Confidence interval:
The mean of Group One minus Group Two equals 0.033
95% confidence interval of this difference: From -1.564 to 1.630
Intermediate values used in calculations:
t = 0.0898
df = 2
standard error of difference = 0.371
Is this a legitimate experiment or would I use a different statistical method?
I must admit that I am confused on the capability of the Paired T test. Please let me know if I am using it correctly, question is a little detailed but I want to make sure that I understand it:
Say you are you have a group of teachers and a group of students.
The Teachers instructs during the week to traditional high school students and on the weekend to GED students---The subjects are exactly the same that the Teachers cover to the two groups yet the teacher instruct different courses, so the curriculum is exactly the same.
The teachers at the end of the year give their students an opportunity to evaluate the students experience regarding the lessons—giving the student let’s say the ability to rate the difficulty level of the subject matter, content, and assignment load.
The group of teachers wants to see if the difficulty level of the curriculum is equal between the two groups. So they add up the scores of the evaluations. Adding all the scores of the evaluation together to get the scores below (so you have a group of teachers who teach different topics then there fellow teachers but the topics they teach are mirrored in High School and the GED class’s (i.e. the history teacher teaches the same history class to the GED students and the High School students, math teacher and so on) For they are not worried about the class’s as individuals but rather the comparison of GED class’s vs. High School class’s. The goal would be to find if the students in the 2 different groups view the material the same way, given the fact that they are 2 different groups but are being taught the same material by the same teacher.
An example of the math
Subject matter
GED= 10.4 High School=11.1
Content
GED=11 High School=10.5
Assignment Load
GED=13.1 High School=12.8
Group 1 (GED)
10.4, 11, 13.1
Group 2 (HS)
11.1, 10.5, 12.8
If you ran the data you would get:
Paired t test results
P value and statistical significance:
The two-tailed P value equals 0.9366
By conventional criteria; this difference is considered to be not statistically significant.
Confidence interval:
The mean of Group One minus Group Two equals 0.033
95% confidence interval of this difference: From -1.564 to 1.630
Intermediate values used in calculations:
t = 0.0898
df = 2
standard error of difference = 0.371
Is this a legitimate experiment or would I use a different statistical method?