Jump to content

User:Leolsz/Alternative hypothesis/Shg7D1 Peer Review

From Wikipedia, the free encyclopedia

General info[edit]

Whose work are you reviewing?

Leolz

Link to draft you're reviewing
https://en.wikipedia.org/wiki/User:Leolsz/Alternative_hypothesis?veaction=edit&preload=Template%3ADashboard.wikiedu.org_draft_template
Link to the current version of the article (if it exists)
Alternative hypothesis

Evaluate the drafted changes[edit]

The biggest shortcomings that I see with this article are issues with grammar, clarity, a lack of examples, and a lack of citations. Here are a few suggestions for improvement:

  • There are a number of grammatical and clarity errors in the lead.
    • The first sentence is both unclear and grammatically incorrect. "a position that states something is happening" is both inaccurate and unclear, and the alternative hypothesis is not necessarily "preferred" over the original hypothesis (e.g. testing whether or not gender affects test scores).
    • “In statistics, alternative hypothesis is…” should be changed to "In statistics, the alternative hypothesis is…”
    • The last sentence, "In the domain of inferential statistics two rival hypotheses can be compared by explanatory power and predictive power," is vague and some elaboration as to why we might find this useful would be nice.
  • The term research hypothesis is not clearly defined, and there are no citations to show that this term is in fact used in academia.
  • The example is very bare-bones. It would be helpful to explain that the test assumes the null hypothesis to be true and determines how likely it is we see our observed data if the null hypothesis is true, in order to determine if the alternative hypothesis is false. This shows that the alternative hypothesis represents a departure from what we expect/assume
  • The section for types of hypothesis contains no citations
  • Point hypothesis could be better explained
  • The explanations for both one-tailed and two-tailed hypothesis reference a distributions tail, but it may be unclear what "tail" refers to (assume that someone with no knowledge of statistics is reading this article. How can we add relevant information or direct them to a page that includes said information?)
  • The article could benefit from more links to other articles and/or more graphics, e.g. that of the critical region for a one-tailed versus a two-tailed test
  • It would probably be helpful to include the mathematical formulation for different alternative hypothesis: e.g. if H_0 is \mu = 0, then a one-tailed alternative hypothesis could be that \mu > 0 and the two-tailed alternative hypothesis is \mu != 0.
  • The difference between a non-directional and a two-tailed alternative hypothesis is unclear to me. It might be useful to give an example of a non-directional alternative hypothesis that isn’t also a two-tailed alternative hypothesis