10th International
Conference on
Teaching Statistics
8 – 13 July 2018
Kyoto, Japan
Contributed paper list


Contributed Paper C187

 

Improving the teaching of basic statistical inference to social science students


Presenter

Hendry Raharjo (Sweden)

Abstract

The teaching approach is based on participant-centered learning of 60-minute workshop. The class starts with a real problem for the students to estimate the true proportion of red beads in a box containing approximately 4000 beads by random sampling. Each student draws his/her own sample two times of which the second sample size is doubled. A step-by-step guide along with how to compute p-value and confidence interval using JMP software is provided. Think-pair-share method is used for interpreting p-value, especially when there is lack of significance. Confidence interval (CI) interpretation is done by showing entire class’ results which come from the same box (population). Changing the null hypothesis is suggested to compare p-value and CI results. The same analysis is repeated using the larger sample.