Talk:Statistics education

New Title

I would suggest the title, "Teaching Statistics in the United States". Kiefer.Wolfowitz (talk) 11:02, 8 November 2009 (UTC)

I would suggest expansion in several directions, including coverage of statistics education in schoools, in non-maths courses in universities, in professional training for other professions, in general education of the public, and of course to other countries. The present major portion of content about whether statistics should be taught by statisticians should not be the major topic of the article. Melcombe (talk) 10:39, 9 November 2009 (UTC)

You, sir, are correct! Kiefer.Wolfowitz (talk) 20:24, 20 January 2010 (UTC)

Updated lead

The term "statistics education" refers to not only the actual teaching & learning, but also the research discipline. See mathematics education and physics education for similar examples. I've updated the lead to make this more clear. Statisfactions (talk) 17:52, 26 October 2012 (UTC)

A paragraph with some problems

I suggest that the second paragraph of the lead needs some work. Taking up the sentences in turn:

Statistics is both a formal science and a practical theory of scientific inquiry, and both aspects are considered in statistics education.

The claim here is that statistics is a theory of scientific inquiry. Many such theories have been proposed in the past. Some, like Bacon's Novum Organum, have been prescriptive, laying out what scientists ought to do; others, like Kuhn's The Structure of Scientific Revolutions, are descriptive accounts of how science actually proceeds. Does statistics participate in this tradition? Was Kuhn remiss in failing to discuss prior work in the field by statisticians? I think not. Statistics does assist in the design of some experiments and is used to judge the adequacy of the evidence for some scientific claims, but this prescriptive function falls far short of providing "a theory of scientific inquiry".

Education in statistics has similar concerns as does education in other mathematical sciences, like logic, mathematics, and computer science.

The first time through this sentence, I thought it was saying that education in statistics has concerns similar to those of statistics, the subject of the prior sentence, adding that other educational endeavors share these concerns. Is this the claim? Comparing the leads of the Statistics and Statistics education articles, it seems that statistics is concerned with the collection, organization, analysis, interpretation, and presentation of data while statistics education deals with the teaching and learning of statistics. While teaching and learning can and should be studied statistically, these do seem to be quite separate concerns.

The fifth time through this sentence, I thought that it might mean that education in statistics has concerns similar to those of education in mathematics and logic. The section Statistics education#Teaching methods, however, seems to imply that it is disputed whether statistics education should follow the traditional pattern of other types of mathematical education. Doesn't this mean that, according to some, the concerns are different? If a thesis is represented as controversial in the body of an article, it should not be stated without qualification in the lead.

At the same time, statistics is concerned with evidence-based reasoning, particularly with the analysis of data. Therefore education in statistics has strong similarities to education in empirical disciplines like psychology and chemistry, in which education is closely tied to "hands-on" experimentation.

This does not follow. Similarity in the subject matter of two fields (doubtful in this case) entails neither similarity in the methods of investigation nor similarity in teaching or learning methodology.

Peter Brown (talk) 21:49, 27 October 2012 (UTC)

I agree that paragraph could use some work. The larger point of that paragraph is to distinguish the two main aspects of statistics, which each have consequences for the goals of statistics education. There's the formal mathematical theory, and there's the practice of actually analyzing data and interpreting results in a variety of contexts. Skills of learning mathematics are relevant for understanding the formal mathematical theory, which is why statistics education shares elements of mathematics education. Statistical investigation, in turn, overlaps with a variety of discipline-specific investigation techniques (economics, biology, psychology) and so the ways people have found to teaching good data-analysis skills, habits, and processes are also relevant to the teaching and learning of statistics.

I agree that the phrasing "practical theory of scientific inquiry" is problematic. Something like "empirical practice" may be a more useful phrasing.

I believe your fifth-time reading is correct -- educators in statistics have concerns similar to educators in mathematics and logic. However, the whole point of the paragraph is to say that it has some concerns in common with math education, but some other concerns in common with chemistry and other empirical disciplines. I don't think it's particularly controversial that statistics education has some concerns in common with math education, since statistics includes a fair amount of math; the controversy is whether statistics should simply be taught as a branch of math, which the lead makes no claims about.

As for the bit about whether the content areas have much in common, I think the claim is that the methods of investigation between statistical inquiry and non-statistical empirical inquiry are similar. This seems plausible to me, but I agree that it's inadequately supported.

This paragraph illustrates the difficulty in defining statistical education: much of the turf of statistics overlaps with that of many other disciplines; for instance, I took an "Econometrics" class which was specialized techniques for analyzing economics data that drew on a combination of economic and statistical theory. Was that class statistics education? Economics education? This is part of the nature of statistics; as a paper by statisticians George Cobb and David Moore put it, it is the "interplay of pattern and context" and can be difficult to separate out distinctly.

Statisfactions (talk) 02:47, 30 October 2012 (UTC)

One could make a vast improvement in the paragraph simply by using your statements with a bit of rewording! How about:

Statistics is a formal mathematical theory as well as an approach to analyzing data and interpreting results in a variety of contexts, and each has consequences for the goals of statistics education. Since statistics involves a considerable amount of mathematical manipulation, education in the field needs to utilize techniques appropriate in mathematical education; statistical investigation overlaps with a variety of discipline-specific investigation techniques, however, and many feel that other methods are also needed. As statistical and nonstatistical empirical inquiry are similar, the respective educational approaches have much in common.

I do suggest that the controversy needs to be alluded to. According to WP:LEAD, "The lead should. . .summarize the most important points—including any prominent controversies." Otherwise, readers will continue into the body of the article thinking that the concerns (not just some concerns) of statistics and mathematics education are much the same.

There is much here that needs further elaboration, best done in the first body section. Going straight into national approaches to the subject is premature.

Peter Brown (talk) 15:22, 30 October 2012 (UTC)

Proposed Removal of "Rarity of Randomized Experiments" subsection

The "Rarity of Randomized Experiments" subsection discusses the rarity of randomized experiments in mathematics education.

I propose removing this section, since this article is on statistics education. The article could certainly use more detail on the methodology of statistics education research, but criticism of math education methodology seems off the mark. Also, the math education article already has a section covering this topic, so I don't think moving the section to that article makes sense.

Statisfactions (talk) 19:30, 12 April 2013 (UTC)

Now done. See the old section if you're curious about what was deleted. Statisfactions (talk) 03:15, 17 April 2013 (UTC)

Suggested additions from the US Conference on Teaching Statistics

A group of statistics educators met at the US Conference on Teaching Statistics to discuss this article. Here are some of the things that people suggested including:

• Why stat ed is not the same as math ed
• Giving a sense of statistics education at all levels (based on national standards for statistics instruction, where relevant)
• Including the history of statistics education
• Relation between current state of statistics and current state of statistics education
• Motivations for statistics education
• For the "programs" section, include groups that are interested in stat ed even if they don't call themselves a "stat ed" program (e.g. NCSU)
• Discuss some controversies in statistics education
  • Use of t-tables
  • Use of graphing calculators
  • resampling vs. asymptotic approaches
  • conceptual vs. computation
  • Bayesian vs. Frequentist
  • How much probability to include
  • Empirical vs. simulated probability distributions

Feel free to jump in and discuss any of these suggestions, or add them to the article! Statisfactions (talk) 14:35, 24 May 2013 (UTC)