Catch up on stories from the past week (and beyond) at the Slashdot story archive

 



Forgot your password?
typodupeerror

Comment: Re:Three puzzles (Score 1) 208

by fropenn (#49495719) Attached to: Social Science Journal 'Bans' Use of p-values
Do you really want someone to answer, or are these all rhetorical?
Here's my take on this issue: Just because something is prone to be misused and misinterpreted doesn't mean it should be banned. In fact, some of the replacement approaches use the very same logic just with a different mathematical calculation process. However, it does illustrate the need for researchers to clearly communicate their results in ways that are less likely to be misused or misinterpreted. This wouldn't exclude the use of p-values but they are only one of many possible tools for researchers to use.

Comment: Re:Hurrah for sex-segregation! (Score 1) 599

Some research has suggested benefits for same-sex segregated education, particularly for girls. http://files.eric.ed.gov/fullt... The separate but equal issue is a problematic, but there may be ways to reproduce the benefits of same-sex education without the full separation (such as dividing a class into two groups, or having a co-educational school with single-sex classes). Sex segregated schools can also be problematic for the transgendered, but at younger ages I don't think it would be as big of a deal for most individuals.

Comment: Re:Statistical Practitioners need to Modernize (Score 1) 115

by fropenn (#47765609) Attached to: Statistics Losing Ground To CS, Losing Image Among Students
I would add that many disciplines are recognizing the importance of statistics and are therefore introducing applied statistics courses for [discipline X]. This causes a drop in enrollment in the pure statistics courses, thus decreasing the number of pure statistics instructors, thus decreasing the demand for individuals trained in pure statistics. In this way statistics is losing itself as a discipline and is quickly becoming specialized into various disciplines (e.g., the application of statistics for medical research).

Make it right before you make it faster.

Working...