Sample size and calculation requirements of t tests Your book discussed ideal sample size as well as how to calculate a t test by hand. In this activity, you will remind yourself how the hand...

1 answer below »




Sample size and calculation requirements of


t


tests





Your book discussed ideal sample size as well as how to calculate a
t
test by hand. In this activity, you will remind yourself how the hand calculation works and how the required calculation is related to your data points.



  1. How could you calculate a
    t
    test if you were interested in change over time?

  2. How could you calculate a
    t
    test if you were using a matched pairs design, using the variables already present? (i.e., gender).

  3. Let’s say that you had a sample of 100 volunteers (you did not have access to the database identified on the picture). You are looking for husband–wife pairs.


    1. 30 participants say they are not married.

    2. 20 participants say they are in homosexual relationships.




  • 30 participants said their spouse did not want to participate.



  1. The remaining participants meet the eligibility requirements.



  1. Given how the dependent samples
    t
    test is calculated, how many participants do you have left?


Answered Same DayJul 09, 2021

Answer To: Sample size and calculation requirements of t tests Your book discussed ideal sample size as well as...

Rajeswari answered on Jul 10 2021
115 Votes
61654 Assignment
1. How could you calculate a t test if you were interested in change over time?
When we wan
t to calculate a t test for change over time, we use correlation coefficient. We treat time as independent variable as it would be moving as it is, and the variable we have for various times as dependent variable. We try to find if there is any correlation between the time and variable we use. For this we find out the correlation coefficient between time x independent variable and y our variable as dependent variable.
Correlation coefficient always lies between -1 and +1. If absolute value is nearer to 1, then there is strong correlation and if nearer to 0 we can say there is almost nil or very weak correlation. Sign of r gives the effect on y. If r is negative, then y decreases as x increases and vice versa.
To check correlation coefficient
H0: r=0
Ha: r≠0
(A two tailed test at significance level say 5% is conducted)
Test statistic = t =
Degree of freedom = n-2
If t statistic lies within critical values we accept r=0 or there is no association between x and y. ...
SOLUTION.PDF

Answer To This Question Is Available To Download

Related Questions & Answers

More Questions »

Submit New Assignment

Copy and Paste Your Assignment Here