A recent publication:
Anita Bright. Education for Whom? Word Problems as Carriers of Cultural Values. Taboo: The Journal of Culture and Education. (Spring 2016). pp. 6-22. Link No paywall.
observes that many textbook math word problems are from "middle or upper class perspectives". The first example word problem is:

Two art students are touring Paris. They each buy a one-day museum pass for \$14. Each student also buys a ticket to the Eiffel Tower for \$11 and a boat ticket for \$3. How much do the two students spend altogether? Explain.

From my reading of this paper, there is an implicit claim that these word problems (with middle or upper class perspectives) have a negative impact on the learning of some less-privileged students.
Is there evidence for this claim ?

The hard evidence could be a study similar to those in the answers to a similar question about gender. There could be comparisons of student performance on different variations of the same word problem, with the same math calculations but with different characters and situations.

The soft evidence could be anectodal, perhaps something like this:

  • a student is stuck on the term "one-day museum pass"
  • the teacher suggests the student replace it by "movie ticket"
  • the student now solves the problem

Thank you for your replies.

  • 3
    $\begingroup$ I'm careful to use very generic clothing, sneakers, socks, etc, or foods, hamburger, fries, drink, etc, in my examples. I understand the vocabulary we use can impact understanding, interesting question. $\endgroup$ Commented Jul 10, 2016 at 2:52
  • 4
    $\begingroup$ @JoeTaxpayer What if the student is vegan and is upset about the mention of hamburgers? There is no such thing as generic culture. $\endgroup$ Commented Jul 10, 2016 at 12:44
  • 1
    $\begingroup$ @JohnColeman - you can extrapolate that to imply that there's no safe words at all. The sandlot wearing student who can't afford socks, the student whose family walks to work as they can't afford a car, etc. are you even comfortable considering 'movie' safe? $\endgroup$ Commented Jul 10, 2016 at 13:36
  • 2
    $\begingroup$ @JoeTaxpayer Almost by definition a word problem involves nonmathematical words, words which have a cultural context. Part of becoming educated is being able to deal with words outside of your immediate cultural experience. I've never played poker in my life and have never gambled, but don't find anything particularly difficult about discussing poker. I'm not convinced that students (weak or otherwise) need to be sheltered from descriptions of lives other than their own. $\endgroup$ Commented Jul 10, 2016 at 16:21
  • 1
    $\begingroup$ I'm not convinced that there is a need to try for a least common denominator culture. Obviously there is nothing to be gained in using examples that contain terms that the students don't understand. But the question seems to go beyond that and suggests that questions shouldn't be used if the students can't relate to the topic, which seems too extreme to me. I have no experience playing tennis and regard it is an upper-class sport that was emphatically not part of my cultural background, but I see no reason why I shouldn't be able to answer questions about the dimensions of a tennis court. $\endgroup$ Commented Jul 11, 2016 at 1:53

4 Answers 4


Yes, there is evidence for the claim; for example, consult the following:

Abedi, J., & Lord, C. (2001). The language factor in mathematics tests. Applied Measurement in Education, 14(3), 219-234. Link (no paywall).

You will find in this piece that the authors, as you considered in your "soft evidence" proposal, modified mathematical questions (from the NAEP) and checked to see the impact on student understanding.

Here are two images; I will mark some relevant components in red.

The first image is of the title and abstract:

enter image description here

The second image is of the beginning of the Discussion section (PDF 13/17):

enter image description here

There are additional subtleties to answering your question; for example, consider the intersection of socioeconomic status and the language spoken by students at home, and you may see that ELLs (cf. bullet-point 1 above) are disproportionately low-SES.

The suggested article may have other references of interest; as far as subsequent articles that cited this piece, here is a link to google scholar's Cited by (with the results cut down by searching for the term "math"). I cannot think of a "seminal article" offhand that addresses your question (although there may well be one!) but I am quite confident (for scholarly and anecdotal reasons) that the answer is, yes, privilege (as measured e.g. by some SES metric) will be significant when looking at student performance on certain types of word problems.

  • 2
    $\begingroup$ While this is interesting it seems to be talking about linguistic complexity rather than class privilege, so it doesn't strike me as directly related. $\endgroup$ Commented Jul 10, 2016 at 12:48
  • 2
    $\begingroup$ I agree that this study is not directly related. I was hoping that the word problem modifications would substitute characters and situations that change the middle or upper class perspective to one that is more familiar to the students. I understand that these sorts of modifications are more difficult, so I accept this answer with the hope for a better one later. The title of a more relevant study might be similar to The Class Perspective Factor in Mathematics Tests. $\endgroup$ Commented Jul 10, 2016 at 23:54

I only have anecdotal evidence but since you said it was welcome, I am adding my 2 cents from both the perspective of a teacher and an item writer, hired to write questions for standardized tests.

I taught math to gifted elementary students. A small fraction of them were gifted in math and had reading levels below grade level. Students with poor reading levels often have poor vocabularies and have a harder time with unfamiliar words.

These students felt safe asking me about the meaning and context of all sorts of words that made no sense to them even on tests. Before I instituted this policy, these students did very poorly in word problems.

An older student or adult might be able to solve a problem without knowing all the words, but younger students get stuck on word meanings and can't seem to move on the problem until they can read it through and know the meaning of all the words.

I noticed in class that students were often stuck on multi-cultural names and words, such as unfamiliar foods. Names were a particular problem if there was a girl or a boy, because the student wouldn't know which was he or she.

When I wrote questions for standardized tests, I was given instructions about wording from different companies. I was told to be sensitive to gender and that no one in my problems could have a gender; telephone repairman became telephone technicians and instead of Emma and Jake, there were just two students. I was told to be sensitive to students who were poor; brunch became lunch because poor students didn't have brunch. I was told to use very simple sentence structure and language.

When I wrote questions for my own students, I used their names and hobbies. I found this delighted them and piqued their interest. If a student needed a makeup assignment, I wrote 6 or 7 problems with just that student's name, which was always a hit.

From my experience language and context in word problems can motivate and delight students or frustrate them and put obstacles in their path.

  • 3
    $\begingroup$ From Amy's and @Benjamin's answers I am coming to the conclusion that any claimed effect of class privilege can truly be traced to language and/or vocabulary. The observation that children cannot move on a word problem until they understand all the words is new and surprising to me. I now think a formal study titled The Class Perspective Factor in Mathematics Tests is impossible to do. It is also impossible to write word problems with a common vocabulary understood by all students, but common words are preferred. $\endgroup$ Commented Jul 28, 2016 at 0:55
  • 1
    $\begingroup$ I worked with gifted children who were perfectionists. Perhaps other children would be willing to tackle word problems without understanding ever word. $\endgroup$
    – Amy B
    Commented Jul 28, 2016 at 4:45

First, the hard evidence is obvious -- our standardized mathematics assessments have continued to show a discrepancy between race and class. This should tell us there is something problematic about the assessment, however usually people who talk the test read this as problems with students.

Second, I feel Ben Dickman's suggestion is completely valid, however, for a different taste, you might want to check out these examples, and follow the larger threads you'll find within. There is a massive literature of critical mathematics education:

  1. Valerie Walkerdine- The Mastery of Reason - basically redid Piaget's experiments and notes the problems with the way these investigations took place in light of student difficulty

  2. Robyn Zevenbergen - Constructivism as a Liberal Bourgeois Discourse or Technologizing Numeracy: Intergenerational Differences in Working Mathematically in New Times

  3. Robert Moses - Radical Equations, discusses some alternative ideas for engaging students with contexts familiar to their own lives.

  4. Jean Lave(all) - Investigates individuals working on mathematics problems in different contexts; situational cognition.

  • $\begingroup$ "First, the hard evidence is obvious -- our standardized mathematics assessments have continued to show a discrepancy between race and class." There could be other reasons. And if it is obvious, someone probably has studied it or more specific, less obvious, version of it. Are your links to empirical studies that test the claim in the question? $\endgroup$
    – Tommi
    Commented Jan 13, 2018 at 19:31
  • $\begingroup$ These are all academic writers, and there is much literature on class and race bias in science and mathematics education. These could be an entry into some of this literature. What other reasons do you propose? $\endgroup$
    – jfkoehler
    Commented Jan 13, 2018 at 19:38
  • $\begingroup$ The writers are academic, but do they provide hard evidence? Teacher bias is an example of an alternate explanation for differences in performance, and it is fairly distinct from word choice in word problems causing difference in performance. $\endgroup$
    – Tommi
    Commented Jan 14, 2018 at 8:34

I began teaching math when SMSG just came in. I mention this for background only (and as a clue to my age!). Before this and to this time, I am sure, word problems were always the nemesis for students. I always suggested (and practiced) this philosophy.
"Word problems are easier to solve after you turn them into math. The problem is that you need to strip away the "garbage". As teachers you (and your students) should practice this and then you and they will not have to deal with PC (politically correct) and in the process will eliminate what may be perceived PC factors. With:

"Two art students are touring Paris. They each buy a one-day museum pass for \$14. Each student also buys a ticket to the Eiffel Tower for \$11 and a boat ticket for $3. How much do the two students spend altogether? Explain."

After reading this You might say, What is important (to the solution) about: "Two art students are touring Paris." Paris? Touring? students? 2? etc.. continue this and perhaps you/they get to the short version:

2(14 + 11 + 3) or something like this.

Then you might ask them to write the word problem that corresponds to this: 2(6 + 17) - 3 and have them read it.

  • 4
    $\begingroup$ This doesn't answer the question. $\endgroup$
    – user507
    Commented Jul 12, 2016 at 17:04
  • $\begingroup$ Also, what is SMSG? $\endgroup$
    – Tommi
    Commented Jan 13, 2018 at 19:34

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.