Statistical/practical significant difference

Forums:
Email this Topic • Print this Page

Hi
Suppose that we have two groups that they select N items. For example, in the table below, we have the number of people from each group that have selected each item.
items: A B C D E F G H K L
G1: 3 4 52 633 1000 0 6 915 890 500
G2: 1 19 17 333 1000 25 6 110 951 510
Population of group 1: 1000
Population of group 2: 1050

The populations are free to select as many of the items as they want. We do not have/know any assumption for selection of items. I want to find which items are important for discrimination between two populations by some test. Is there any test that determines the differences in which items are statistically/practically important/significant?
Thanks very much for your guidance.
Jaafar

Topic Stats
Top Replies
Link to this Topic

Type: Question • Score: 1 • Views: 266 • Replies: 4

[+3] - engineer - 10/12/2020Hi Jaafar. First, find the probability that an item will be chosen. Let's look at item D. 633 samples were chosen out of 4003 total selections for Group 1 or 15.8%. Now find the confidence...

@Jaafar,

Hi Jaafar.

First, find the probability that an item will be chosen. Let's look at item D. 633 samples were chosen out of 4003 total selections for Group 1 or 15.8%. Now find the confidence interval around that. The 95% confidence interval is 1.97*sqrt( P x (1-P) / N ) or 1.97 * sqrt ( .158 * .842 / 4003) = 1.1%. Combined, we can say the probability that someone in group one would use a choice to select item D is 14.7% to 16.9%. If you do the same for group 2, you get 11.2% chance of selects with a 1.1% confidence interval or 10.1% to 12.3%. Those two ranges do not overlap, so you can say they are statistically different.

1 Reply

@engineer,

Hi engineer,

Thank you for your answer. Would you please tell me the reference source or paper of this method? Thanks!

1 Reply

@Jaafar,

https://en.wikipedia.org/wiki/Binomial_proportion_confidence_interval#:~:text=In%20statistics%2C%20a%20binomial%20proportion,failure%20experiments%20(Bernoulli%20trials).&text=The%20observed%20binomial%20proportion%20is,turn%20out%20to%20be%20heads.

Statistical/practical significant difference

Quick Links

My Account

able2know