This is the
talk page for discussing improvements to the
Statistics template. |
|
Statistics Template‑class | |||||||
|
Should the template include notable statisticians? I think it may be too much, but what does everybody else think? Lilac Soul ( talk • contribs • count) 08:23, 28 August 2007 (UTC)
Could topics like Bimodal distribution and Binomial distribution be included? They seem to be already included in the ProbDistribution navbox, but that navbox is very large. Rhetth ( talk) 22:41, 4 August 2008 (UTC)
Apparently, links to Wikipedia:WikiProject_Statistics in this template violates WP:SELF - WikiProjects should not be included in articles or on their templates. I have removed.— G716 < T· C> 01:13, 22 September 2008 (UTC)
This template has become quite large, and as WP:NAV says, "a large template has limited navigation value". I suggest splitting it into several smaller templates on each sub-topic. (We could draft them as sub-pages of this talk page to start with, e.g. Template talk:Statistics/Experimental design.) We could also consider a "Major topics in statistics" navbox to give an overview, in the same style as e.g. Template:Number theory-footer, Template:Applied-footer etc. Thoughts / volunteers? (I'm not taking this on on my own..) Qwfp ( talk) 10:22, 1 October 2008 (UTC)
I have removed publications and history sections to reduce size of template. If anyone objects, please revert. — G716 < T· C> 12:20, 16 April 2009 (UTC)
Perhaps have a separate template for the subset of statistics that pertains to econometrics? cancan101 ( talk) 02:19, 12 March 2009 (UTC)
Following the previous discussion, I augmented the template on design of experiments, resulting in the operational template Template:Experimental design. The template is designed for (somewhat) more knowledgable users, than is the basic statistics template. For example, Student's t is listed next to Hotelling's T, to provide context for the latter.
A few Bayesian entries occur at the end of the lines (which provides some vertical organization, also)
The horizontal line between inference and designs is awkward, but should aid the reader (imho). (Maybe somebody knows a better solution?)
Feedback now is especially welcome.
Kiefer.Wolfowitz (
talk) 11:49, 23 June 2009 (UTC)
Any thoughts on the relevance of the first three items under this heading: Null hypothesis • Alternative hypothesis • Type I and Type II errors • Statistical power • Effect size. I would suggest dropping them entirely as they aren't needed elsewhere in a brief list of topics. Melcombe ( talk) 15:55, 9 December 2008 (UTC)
The option {{Statistics|state=collapsed}} appears not to work .... can it be fixed? Melcombe ( talk) 10:21, 2 December 2009 (UTC)
I doubt that volcano plots deserve to listed on this template, and am concerned that they were added without prior discussion. I have asked the responsible editor to consider removing this item, and to discuss the importance of such plots here before adding them. Thanks! Kiefer.Wolfowitz ( talk) 22:15, 3 January 2010 (UTC)
I comment on imperfections: "Non-standard" is wrong because the Gauss-Markov theorem justifies least squares.
Instead, I would suggest something like
Thanks. Kiefer.Wolfowitz ( talk) 19:00, 21 January 2010 (UTC)
I'd like to see a section for exact tests added to the template, including the binomial test, multinomial test, Fisher's exact test, Barnard's test, and any others I forgot. Drdan14 ( talk) 01:01, 8 March 2010 (UTC)
Natural experiment and observational study---these two topics are important in discussions of empirical science and especially social sciences like economics and epidemiology. I moved both of them to the back of the experimental section, since they are usually discussed as methods that are used when randomized methods are too expensive, unethical, etc. I added "randomization", which is usually taught as a fundamental concept in statistics (Peirce, Fisher, Rao, Moore McCabe, Freedman, ASA guidelines for a first course, etc.). I moved the sampling-topics to the same section on data-collection, leaving questionaires, etc. in the social science section below. Thanks, Kiefer.Wolfowitz ( talk) 14:44, 30 April 2010 (UTC)
I would suggest moving survey topics back to social-statistics. There is now duplication in social-statistics and sampling. Thanks! Kiefer.Wolfowitz ( talk) 01:30, 8 May 2010 (UTC)
Since i suggested above this template had got too big 18 months ago it has very nearly doubled in size! It's now too big to fit in the browser window on my laptop screen (bigger than some) without scrolling, and that certainly limits its navigational value. If we gave it a severe pruning i can't see any easy way to suppress regrowth, so i'd like to again suggest splitting it up. Most of the existing (main, i.e. leftmost) groups would be suitable for separate navboxes. Thoughts? Volunteers? (I'll volunteer to take on Survival analysis myself this time.) Qwfp ( talk) 20:57, 25 May 2010 (UTC)
Two suggestions, which would have the topical ordering follow the ordering of elemtentary textbooks in the USA and Sweden:
First, move "data collection" up, before "inference".
Second, have the "data analysis" follow "regression". (The "data analysis" could be renamed "multivariate dependent random variables", also.)
(This is not a big issue, imho.).
My previous re-ordering of data collection and inference was effectively reverted by editor Melcombe. No other editors commented on my previous re-ordering. New blood or fresh arguments might move the ordering towards consensus! Thanks, Kiefer.Wolfowitz ( talk) 23:07, 6 June 2010 (UTC)
I have un-wikilinked "study design", and hidden "regression discontinuity", on a possibly temporary basis. The object here is to see what other articles link to them, and this is effectively impossible while they are in this template. It seems to take about a week for the effect of this type of removal to feed through to all of Wikipedia's underlying cataloguing which is used for the "what link's here" tool. Melcombe ( talk) 14:49, 7 July 2010 (UTC)
I have replaced the previous group name of "Data analysis" by "Data analyses and models for particular data types", partly to reflect concerns expressed above, but also as a trial of using a longer name to see if this length is acceptable to those concerned with using up too much space. Melcombe ( talk) 09:32, 14 July 2010 (UTC)
I propose moving the line headed "Correlation" from under "Correlation and regression" to be under "descriptive statistics" as correlation seems to be more a descriptive thing rather than a modelling activity. Alternatively, since lines are mostly hidden when displayed, it could be under both group headings (possibly with slightly different content). Melcombe ( talk) 09:40, 14 July 2010 (UTC)
Following the above comment about "categorical data", I have added a "Categorical data" group under analyses. I think the group under "descriptive" was meant to be about data that had been summarised into something like contingency tables, rather than analyses of contingency tables. So I have revised the groups under "descriptive" to try to be clearer. Melcombe ( talk) 11:33, 15 July 2010 (UTC)
Estimation is obviously an inferential procedure.
Some sampling-theory estimators are listed in the grouping of procedures for "frequentist" (sic., i.e. sampling-theory) inference, which shows some inconsistencies with the present grouping. I suggest merging the "general estimators" into the "frequentist" inference grouping. Thanks, Kiefer.Wolfowitz ( talk) 17:39, 26 October 2010 (UTC)
Dear Melcombe and other editors,
Moving "McNemar's test & Cohen's kappa" from their position in "Data analyses and models for other specific data types" would leave that category solely to have stochastic processes or multivariate data: Then the category could be renamed " Multivariate & Stochastic Data". Would that be acceptable?
Best regards, Kiefer.Wolfowitz ( talk) 18:58, 15 November 2010 (UTC)
I removed " errors and residuals" as a group heading because the article contents contain nothing at all about ANOVA etc.. Similarly, Goodness of fit contains nothing much at all about regression, but is much more general in content. And Deviance (statistics) is arguably too technical for a general-scope nav-box. After all, we can't have all 2000 stats articles appearing. But I did create a new general-regression sub-group and placed " errors and residuals" within it. Melcombe ( talk) 10:43, 16 November 2010 (UTC)
Analysis of variance is a generic term, which is traditionally considered in terms of the orthomodular lattice of subspaces of an inner-product space. At least since Per Martin-Lof's work of the early 1970s, there have been extensions in terms of so-called statistical models, such as analysis of deviance, redundancy, etc., which try to perform ANOVA in more general settings, with varying degrees of succes and failure. Before Melcombe reverts again, would he be so kind as to specify one most reliable textbook or invited survey article that organizes this material as "formal analysis" rather than ANOVA. Kiefer.Wolfowitz ( Discussion) 15:06, 8 April 2011 (UTC)
Per
WP:Deviations, we should not be overriding the
MediaWiki:Common.css classes using "style attributes". I attempted to remove the |belowstyle=
, but was reverted. Specifying this coloring seems entirely pointless since "CCCCFF" is the default used by
MediaWiki:Common.css, and hence, there is no reason to specify it. Please voice your objections here. Thank you.
Frietjes (
talk) 23:26, 25 April 2011 (UTC)
An editor changed this navbox (along with many others) without getting consensus or asking for comments. This is the core navbox for the statistics project, and major changes should be discussed first, here, with notice given at the project's talk page.
It is bad form, especially for a relatively inexperienced editor, to initiate changes on many navboxes, without discussing things first. There is a talk page for navboxes, and major changes to ~many navboxes generally should be made there first. Kiefer. Wolfowitz 23:31, 25 April 2011 (UTC)
Per
WP:Deviations, we should not be overriding the
MediaWiki:Common.css classes using "style attributes". I attempted to remove the |belowstyle=
, but was reverted. Specifying this coloring seems entirely pointless since "CCCCFF" is the default used by
MediaWiki:Common.css, and hence, there is no reason to specify it. Please voice your objections here. Thank you.
Frietjes (
talk) 1:26 am, Today (UTC+2)
|belowstyle=background: #ccccff
. The reason why that caused a slight change in color is that in
MediaWiki:Common.css the "level 2" color for above and below sections is DDDDFF , which is lighter than CCCCFF . Hence, when this statement was removed it changed the color of the below section from the darker to the lighter color. I see the difference as being very slight, and not something to be worried about. The cool thing about having the color specified in
MediaWiki:Common.css is that (1) anyone can change it by simply putting a new color definition in his/her own
Special:MyPage/skin.css, which includes the visually impaired who might find it useful to increase the level of contrast between the text and the background, (2) the Wikipedia developers can uniformly change the color scheme used by the site, and (3) it ensures that all the navboxes will match each other. My recent edits to statistics navigation boxes were fixing other issues, and were not changes to style or colors. Thanks!
Plastikspork
―Œ(talk) 01:08, 26 April 2011 (UTC)
I commented out the image. I feel as though it was taking up too much space, and is not necessary here. Thank you. Frietjes ( talk) 15:29, 23 May 2011 (UTC)
I've just added a link to scan statistic under Statistical inference > statistical theory. Do feel free to move if you think it would be better elsewhere. Gareth Jones ( talk) 18:08, 6 January 2014 (UTC)
I think some of causal inference articles in Category:Causal inference such as Rubin causal model can be included in this template. Taha ( talk) 07:04, 14 February 2014 (UTC)
I've replaced the link to Statistical population in the Applications/Social statistics section with one to Population statistics, which is the topic that would be relevant to that heading. The article Statistical population, which refers to the population of objects a statistical analysis studies, should probably be linked somewhere in the template; I'm think maybe in Data collection/Study design, or maybe in Statistical inference/Statistical theory. Anyone have a preference? - Bryanrutherford0 ( talk) 18:26, 11 March 2016 (UTC)
A complaint about the overuse and size (14 Kb) of this template appeared in WikiProject Mathematics recently. Statistics is often/usually a separate department and major in American universities, and we don't use a template for history, or for large subfields like American history (see the Missouri Compromise article for how specific their infobox is), and for math the infoboxes are usually subfields like Template:topology or Template:analysis-footer.
I suggest splitting each subsection in the Statistics template to a separate footer or infobox, using those smaller templates in any lower-importance article, while a simplified general Statistics template can still be used for the high-importance general articles. SamuelRiv ( talk) 16:34, 25 August 2016 (UTC)
It seems like a fairly straightforward application of Statistics which is incredibly common in the 21st century and is used in every industry. Particularly image recognition, spatial analysis, deep learning, content personalization, and so many other things... 32.97.110.60 ( talk) 20:36, 20 June 2017 (UTC)
This
edit request has been answered. Set the |answered= or |ans= parameter to no to reactivate your request. |
Add Van der Waerden test to the "Rank statistics sub-subsection of the "Specific tests" subsection of the "Statistical inference" section of the category. B2TF ( talk) 20:40, 7 March 2022 (UTC) B2TF ( talk) 20:40, 7 March 2022 (UTC)
add vector generalized linear model 68.134.243.51 ( talk) 00:19, 22 September 2022 (UTC)
Can we add the Equivalence Test page to the section of "Testing Hypotheses". I think it fits in line with the rest of the subjects. I think it's important to include more hypothesis testing that requires a threshold for practical differences. Treodeo0 ( talk) 23:41, 20 November 2022 (UTC)
This is the
talk page for discussing improvements to the
Statistics template. |
|
Statistics Template‑class | |||||||
|
Should the template include notable statisticians? I think it may be too much, but what does everybody else think? Lilac Soul ( talk • contribs • count) 08:23, 28 August 2007 (UTC)
Could topics like Bimodal distribution and Binomial distribution be included? They seem to be already included in the ProbDistribution navbox, but that navbox is very large. Rhetth ( talk) 22:41, 4 August 2008 (UTC)
Apparently, links to Wikipedia:WikiProject_Statistics in this template violates WP:SELF - WikiProjects should not be included in articles or on their templates. I have removed.— G716 < T· C> 01:13, 22 September 2008 (UTC)
This template has become quite large, and as WP:NAV says, "a large template has limited navigation value". I suggest splitting it into several smaller templates on each sub-topic. (We could draft them as sub-pages of this talk page to start with, e.g. Template talk:Statistics/Experimental design.) We could also consider a "Major topics in statistics" navbox to give an overview, in the same style as e.g. Template:Number theory-footer, Template:Applied-footer etc. Thoughts / volunteers? (I'm not taking this on on my own..) Qwfp ( talk) 10:22, 1 October 2008 (UTC)
I have removed publications and history sections to reduce size of template. If anyone objects, please revert. — G716 < T· C> 12:20, 16 April 2009 (UTC)
Perhaps have a separate template for the subset of statistics that pertains to econometrics? cancan101 ( talk) 02:19, 12 March 2009 (UTC)
Following the previous discussion, I augmented the template on design of experiments, resulting in the operational template Template:Experimental design. The template is designed for (somewhat) more knowledgable users, than is the basic statistics template. For example, Student's t is listed next to Hotelling's T, to provide context for the latter.
A few Bayesian entries occur at the end of the lines (which provides some vertical organization, also)
The horizontal line between inference and designs is awkward, but should aid the reader (imho). (Maybe somebody knows a better solution?)
Feedback now is especially welcome.
Kiefer.Wolfowitz (
talk) 11:49, 23 June 2009 (UTC)
Any thoughts on the relevance of the first three items under this heading: Null hypothesis • Alternative hypothesis • Type I and Type II errors • Statistical power • Effect size. I would suggest dropping them entirely as they aren't needed elsewhere in a brief list of topics. Melcombe ( talk) 15:55, 9 December 2008 (UTC)
The option {{Statistics|state=collapsed}} appears not to work .... can it be fixed? Melcombe ( talk) 10:21, 2 December 2009 (UTC)
I doubt that volcano plots deserve to listed on this template, and am concerned that they were added without prior discussion. I have asked the responsible editor to consider removing this item, and to discuss the importance of such plots here before adding them. Thanks! Kiefer.Wolfowitz ( talk) 22:15, 3 January 2010 (UTC)
I comment on imperfections: "Non-standard" is wrong because the Gauss-Markov theorem justifies least squares.
Instead, I would suggest something like
Thanks. Kiefer.Wolfowitz ( talk) 19:00, 21 January 2010 (UTC)
I'd like to see a section for exact tests added to the template, including the binomial test, multinomial test, Fisher's exact test, Barnard's test, and any others I forgot. Drdan14 ( talk) 01:01, 8 March 2010 (UTC)
Natural experiment and observational study---these two topics are important in discussions of empirical science and especially social sciences like economics and epidemiology. I moved both of them to the back of the experimental section, since they are usually discussed as methods that are used when randomized methods are too expensive, unethical, etc. I added "randomization", which is usually taught as a fundamental concept in statistics (Peirce, Fisher, Rao, Moore McCabe, Freedman, ASA guidelines for a first course, etc.). I moved the sampling-topics to the same section on data-collection, leaving questionaires, etc. in the social science section below. Thanks, Kiefer.Wolfowitz ( talk) 14:44, 30 April 2010 (UTC)
I would suggest moving survey topics back to social-statistics. There is now duplication in social-statistics and sampling. Thanks! Kiefer.Wolfowitz ( talk) 01:30, 8 May 2010 (UTC)
Since i suggested above this template had got too big 18 months ago it has very nearly doubled in size! It's now too big to fit in the browser window on my laptop screen (bigger than some) without scrolling, and that certainly limits its navigational value. If we gave it a severe pruning i can't see any easy way to suppress regrowth, so i'd like to again suggest splitting it up. Most of the existing (main, i.e. leftmost) groups would be suitable for separate navboxes. Thoughts? Volunteers? (I'll volunteer to take on Survival analysis myself this time.) Qwfp ( talk) 20:57, 25 May 2010 (UTC)
Two suggestions, which would have the topical ordering follow the ordering of elemtentary textbooks in the USA and Sweden:
First, move "data collection" up, before "inference".
Second, have the "data analysis" follow "regression". (The "data analysis" could be renamed "multivariate dependent random variables", also.)
(This is not a big issue, imho.).
My previous re-ordering of data collection and inference was effectively reverted by editor Melcombe. No other editors commented on my previous re-ordering. New blood or fresh arguments might move the ordering towards consensus! Thanks, Kiefer.Wolfowitz ( talk) 23:07, 6 June 2010 (UTC)
I have un-wikilinked "study design", and hidden "regression discontinuity", on a possibly temporary basis. The object here is to see what other articles link to them, and this is effectively impossible while they are in this template. It seems to take about a week for the effect of this type of removal to feed through to all of Wikipedia's underlying cataloguing which is used for the "what link's here" tool. Melcombe ( talk) 14:49, 7 July 2010 (UTC)
I have replaced the previous group name of "Data analysis" by "Data analyses and models for particular data types", partly to reflect concerns expressed above, but also as a trial of using a longer name to see if this length is acceptable to those concerned with using up too much space. Melcombe ( talk) 09:32, 14 July 2010 (UTC)
I propose moving the line headed "Correlation" from under "Correlation and regression" to be under "descriptive statistics" as correlation seems to be more a descriptive thing rather than a modelling activity. Alternatively, since lines are mostly hidden when displayed, it could be under both group headings (possibly with slightly different content). Melcombe ( talk) 09:40, 14 July 2010 (UTC)
Following the above comment about "categorical data", I have added a "Categorical data" group under analyses. I think the group under "descriptive" was meant to be about data that had been summarised into something like contingency tables, rather than analyses of contingency tables. So I have revised the groups under "descriptive" to try to be clearer. Melcombe ( talk) 11:33, 15 July 2010 (UTC)
Estimation is obviously an inferential procedure.
Some sampling-theory estimators are listed in the grouping of procedures for "frequentist" (sic., i.e. sampling-theory) inference, which shows some inconsistencies with the present grouping. I suggest merging the "general estimators" into the "frequentist" inference grouping. Thanks, Kiefer.Wolfowitz ( talk) 17:39, 26 October 2010 (UTC)
Dear Melcombe and other editors,
Moving "McNemar's test & Cohen's kappa" from their position in "Data analyses and models for other specific data types" would leave that category solely to have stochastic processes or multivariate data: Then the category could be renamed " Multivariate & Stochastic Data". Would that be acceptable?
Best regards, Kiefer.Wolfowitz ( talk) 18:58, 15 November 2010 (UTC)
I removed " errors and residuals" as a group heading because the article contents contain nothing at all about ANOVA etc.. Similarly, Goodness of fit contains nothing much at all about regression, but is much more general in content. And Deviance (statistics) is arguably too technical for a general-scope nav-box. After all, we can't have all 2000 stats articles appearing. But I did create a new general-regression sub-group and placed " errors and residuals" within it. Melcombe ( talk) 10:43, 16 November 2010 (UTC)
Analysis of variance is a generic term, which is traditionally considered in terms of the orthomodular lattice of subspaces of an inner-product space. At least since Per Martin-Lof's work of the early 1970s, there have been extensions in terms of so-called statistical models, such as analysis of deviance, redundancy, etc., which try to perform ANOVA in more general settings, with varying degrees of succes and failure. Before Melcombe reverts again, would he be so kind as to specify one most reliable textbook or invited survey article that organizes this material as "formal analysis" rather than ANOVA. Kiefer.Wolfowitz ( Discussion) 15:06, 8 April 2011 (UTC)
Per
WP:Deviations, we should not be overriding the
MediaWiki:Common.css classes using "style attributes". I attempted to remove the |belowstyle=
, but was reverted. Specifying this coloring seems entirely pointless since "CCCCFF" is the default used by
MediaWiki:Common.css, and hence, there is no reason to specify it. Please voice your objections here. Thank you.
Frietjes (
talk) 23:26, 25 April 2011 (UTC)
An editor changed this navbox (along with many others) without getting consensus or asking for comments. This is the core navbox for the statistics project, and major changes should be discussed first, here, with notice given at the project's talk page.
It is bad form, especially for a relatively inexperienced editor, to initiate changes on many navboxes, without discussing things first. There is a talk page for navboxes, and major changes to ~many navboxes generally should be made there first. Kiefer. Wolfowitz 23:31, 25 April 2011 (UTC)
Per
WP:Deviations, we should not be overriding the
MediaWiki:Common.css classes using "style attributes". I attempted to remove the |belowstyle=
, but was reverted. Specifying this coloring seems entirely pointless since "CCCCFF" is the default used by
MediaWiki:Common.css, and hence, there is no reason to specify it. Please voice your objections here. Thank you.
Frietjes (
talk) 1:26 am, Today (UTC+2)
|belowstyle=background: #ccccff
. The reason why that caused a slight change in color is that in
MediaWiki:Common.css the "level 2" color for above and below sections is DDDDFF , which is lighter than CCCCFF . Hence, when this statement was removed it changed the color of the below section from the darker to the lighter color. I see the difference as being very slight, and not something to be worried about. The cool thing about having the color specified in
MediaWiki:Common.css is that (1) anyone can change it by simply putting a new color definition in his/her own
Special:MyPage/skin.css, which includes the visually impaired who might find it useful to increase the level of contrast between the text and the background, (2) the Wikipedia developers can uniformly change the color scheme used by the site, and (3) it ensures that all the navboxes will match each other. My recent edits to statistics navigation boxes were fixing other issues, and were not changes to style or colors. Thanks!
Plastikspork
―Œ(talk) 01:08, 26 April 2011 (UTC)
I commented out the image. I feel as though it was taking up too much space, and is not necessary here. Thank you. Frietjes ( talk) 15:29, 23 May 2011 (UTC)
I've just added a link to scan statistic under Statistical inference > statistical theory. Do feel free to move if you think it would be better elsewhere. Gareth Jones ( talk) 18:08, 6 January 2014 (UTC)
I think some of causal inference articles in Category:Causal inference such as Rubin causal model can be included in this template. Taha ( talk) 07:04, 14 February 2014 (UTC)
I've replaced the link to Statistical population in the Applications/Social statistics section with one to Population statistics, which is the topic that would be relevant to that heading. The article Statistical population, which refers to the population of objects a statistical analysis studies, should probably be linked somewhere in the template; I'm think maybe in Data collection/Study design, or maybe in Statistical inference/Statistical theory. Anyone have a preference? - Bryanrutherford0 ( talk) 18:26, 11 March 2016 (UTC)
A complaint about the overuse and size (14 Kb) of this template appeared in WikiProject Mathematics recently. Statistics is often/usually a separate department and major in American universities, and we don't use a template for history, or for large subfields like American history (see the Missouri Compromise article for how specific their infobox is), and for math the infoboxes are usually subfields like Template:topology or Template:analysis-footer.
I suggest splitting each subsection in the Statistics template to a separate footer or infobox, using those smaller templates in any lower-importance article, while a simplified general Statistics template can still be used for the high-importance general articles. SamuelRiv ( talk) 16:34, 25 August 2016 (UTC)
It seems like a fairly straightforward application of Statistics which is incredibly common in the 21st century and is used in every industry. Particularly image recognition, spatial analysis, deep learning, content personalization, and so many other things... 32.97.110.60 ( talk) 20:36, 20 June 2017 (UTC)
This
edit request has been answered. Set the |answered= or |ans= parameter to no to reactivate your request. |
Add Van der Waerden test to the "Rank statistics sub-subsection of the "Specific tests" subsection of the "Statistical inference" section of the category. B2TF ( talk) 20:40, 7 March 2022 (UTC) B2TF ( talk) 20:40, 7 March 2022 (UTC)
add vector generalized linear model 68.134.243.51 ( talk) 00:19, 22 September 2022 (UTC)
Can we add the Equivalence Test page to the section of "Testing Hypotheses". I think it fits in line with the rest of the subjects. I think it's important to include more hypothesis testing that requires a threshold for practical differences. Treodeo0 ( talk) 23:41, 20 November 2022 (UTC)