User survey of Statistics textbook

In MS 150 Statistics this term a new textbook was used, Introduction to Statistics Using OpenOffice.org Calc. The new text was designed with knowledge of our students reading level and vocabulary issues. The text was also based on six years of experience teaching our students statistics. While the language was designed to be more accessible, the text closely parallels in content industry leading texts such as Brase and Brase's Understanding Basic Statistics and Triola's Elementary Statistics using Excel. The intent here was to preserve articulation and content coverage. The chapters in the new text varied in their reading level, with most measuring between ninth and eleventh grade on various measures.

Of 39 students surveyed, 34 (87%) reported buying the text. The remaining 5 (13%) reported having access to a copy of the new textbook "always" (3 students) or "sometimes" (2 students). These five could well be dormitory students. An initial concern at term start was that the textbook was more expensive than initially expected. Only 9 students (23%) felt the text was too expensive. When the question was rephrased "Compared to other textbooks, was the textbook expensive?" only one student said yes. The other 38 students said "no". This leaves open the suggestion that the nine who felt the text was too expensive might object in some absolute sense to any price level above pocket change. There are few texts that can garner a 77% rating of not being too expensive.

The one student who felt the textbook was too expensive compared to other textbooks would still recommend the text to future students and would recommend that they buy the text. This student always found the helpful, never found the text to be useless, and always referred to the text.

Thirty-eight students would recommend the textbook to future students (question 12), thirty-eight students would recommend that future students buy the textbook (question 13). While these two questions seem similar if not redundant, the lone dissenting voice for each question was a different student. The student who dissented on 13 gave a reason, "I think that futures students will not really necessary [sic] to buy the books, because the lectures in class are clear enough and the instructor always post the works in the web page." The later comment is a reference to all the homework assignment being accessible at http://www.comfsm.fm/~dleeling/statistics/statistics.html

Thirty-two students (82%) found the text "always" helpful, twenty-seven (69%) "always" referred to the textbook when doing homework, and twenty-nine (74%) "always" brought the textbook to class. When asked if they could understand the explanations in the textbook, 18 replied "always" and "21 replied "sometimes." No student replied "rarely" or "never." Few would likely argue that a statistics textbook is easy reading, these results are strongly positive results for such a technical text.

Given the above, it is of some interest to note that only nine students report "always" reading the textbook, 28 report reading the textbook "sometimes," and two report "rarely" reading the textbook. As I have learned to my chagrin in SC 130, students tend to not read the textbook. My experience in the course in the past and in other courses suggests that, outside of a course where the textbook is a workbook (such as in developmental math), 37 of 39 reading the text at least "sometimes" is strong reading rate.

When asked "Did you study other statistics book (for example, those in the library)?," six (15%) replied "sometimes," five (13%) replied "rarely," and the remaining 28 replied "never." While the usage rate is low, it is not zero. There is learning value in having supplementary texts in the library, I suspect especially those books that aimed at assisting students having difficulties in statistics.

The last four questions were open response. The students answered these with nothing short of enthusiasm.
16. What part of the textbook was the easiest to understand?
17. What part of the textbook was the hardest to understand?
18. What would you recommend to improve the textbook?
19. What other comments do you have about the textbook (use the back if needed)?

Only four students left these questions blank. While some wrote brief answers, others wrote lengthy notes that were clearly carefully thought out. The notes provide me as the author with a wealth of suggestions on how to improve the text for our students. The ability to alter and adjust the textbook to better fit our students is something that is uniquely possible for a locally authored text.

The student experience with the text has been overwhelmingly positive. Other positives have been a text that was available on class day one. Local publishing is not susceptible to the vagaries of the off-island ordering and shipping process. The text also makes use of local examples and data gathered during the six years of the course.

The table below shows all the complete results for the first 15 questions.

Q Question Yes No

1 Did you buy the texbook? 87% 13%



Always Sometimes Rarely Never
2 If you did not buy the textbook, did you have access to a copy of the textbook? 44% 17% 17% 22%
3 Do you bring your textbook to class with you? 74% 15% 3% 8%
4 Did the instructor refer to the textbook in class? 74% 26% 0% 0%
5 Did you read the textbook? 23% 72% 5% 0%
6 Do you do the homework assignments? 28% 62% 10% 0%
7 Did you refer to the textbook when you did your homework? 69% 28% 3% 0%
8 Was the textbook helpful? 82% 18% 0% 0%
9 Was the textbook useless? 3% 5% 5% 86%
10 Could you understand the explanations in the textbook? 46% 54% 0% 0%
11 Did you study other statistics book (for example, those in the library) 0% 15% 13% 72%


Yes No

12 Would you recommend the textbook to future students of the course? 97% 3%

13 Would you recommend that future students buy the textbook? 97% 3%

14 Was the textbook too expensive? 23% 77%

15 Compared to other textbooks, was the textbook expensive? 3% 97%

January 2008 post-script

I would note that the fall term moved further and through more material than any prior term. I hesitate to credit the text with this impact, but the text's clarity and sharp focus on only that material that we actually cover may have permitted the students to move slightly more rapidly through the material. The spring term syllabus currently has a small amount of slack in the final week of the term, and I am likely to delete section 9.1 after long and careful thought this past break. Chapter nine was cited as the most difficult chapter, with 19 of 39 students (49%) singling out this chapter as problematic. The way 9.1 runs confidence intervals and hypothesis tests (where the population standard deviation is known and the normal distribution holds) is quite different from 9.2 (where the population standard deviation is unknown and the t-distribution must be used), confusingly different.

I would note that I anticipate having a second edition of the statistics text ready to roll out for fall term. The major revisions are actually done - all in pages and pages of scribbled notes and based directly on the student textbook assessment feedback. More products of my beach sabbatical on Kosrae in December 2008. During the spring 2008 term I will be working these changes into the actual text. My target is to have the second edition done by May and thus ready to go to print in May or June 2008.