I am conducting research on academic word lists and ESL textbooks. I have three words lists, each with about 600 or so words on them. I have 12 textbooks that I analyzed and created a word list of words taught. Then I compared each academic word list with each textbook word list to see which words appear both on the academic word list and the textbook word list.

This is what I want to know for my research question: how many words from the text match the words on the list? For example, one textbook had 67 words that also appeared on one academic word list (which had 610 words on it). So, about 10.98 percent of the words in the textbook appear on the academic word list.

I want to know if I can test this for significance--can I test to say a significantly low amount of words from the textbook appear on the word list (only 10.98 appear). Could I do a chi square to get dome data to determine significance? Is there another stat test I could do?

  • $\begingroup$ What exactly is the hypothesis you want to test? What's a "significantly low amount"? It seems like you might be better served by a confidence interval. $\endgroup$ – Dougal Nov 5 '15 at 23:41
  • 1
    $\begingroup$ "low amount" compared to what? $\endgroup$ – Glen_b Nov 6 '15 at 0:19
  • $\begingroup$ For example, if only 10% of the words in the textbook appear on the Academic Word List, and I want to say that 10% is a low amount because I hypothesize that 100percent would appear on the list. Then, I want to be able to say 10percent is significantly lower than the expected 100 percent. Could I statistically analyze my data to be able to show that 10% is significantly lower than the expected 100 percent? Could I create a chi square? (Maybe I can only use the percentage) $\endgroup$ – Jane Nov 7 '15 at 3:41

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.