0
$\begingroup$

We are conducting research to examine a relationship between some triggers and migraine with aura.

the sample obtained by a questionnaire to and the sample classified into two categories (diagnosed with migraine with aura, undiagnosed) and they were asked about there belief of the effect of the triggers (do you notice a relationship between eating hotdog and having an attack of migraine with aura?) I wanted to test whether there is a significant relationship between the trigger and having migraine with aura

In Pearson chi-square, it shows a significant relationship between migraine with aura and (one of the examples) eating hotdog enter image description here enter image description here The problem is the frequency of people who report (no relationship between eating hotdog and having an attack of migraine with aura), is greater than the frequency of people who report that there is a relationship between eating hotdog and having an attack of migraine with aura.

Can someone help me with the interpretation of this result because I'm so confused?

Improve this question
$\endgroup$
3
  • $\begingroup$ Welcome to CrossValidated, Narjes. You will get better (and more prompt) answers if your question is clear. Are you asking about how to interpret the statistical result, or asking for a more appropriate analysis, or asking what your data can tell you about migraine auras? You should probably say more about what the data represent and how the samples were obtained. $\endgroup$
    – Michael Lew
    Commented Dec 20, 2019 at 19:44
  • $\begingroup$ In addition to the suggestions in another comment, it would be helpful to show the two-way table of numbers in addition to (or rather than) the bar chart. $\endgroup$
    – EdM
    Commented Dec 20, 2019 at 20:32
  • $\begingroup$ the sample obtained by a questionnaire to and the sample classified into two categories (diagnosed with migraine with aura, undiagnosed) and they were asked about there belief of the effect of the triggers (do you notice a relationship between eating hotdog and having an attack of migraine with aura?) I wanted to test whether there is a significant relationship between the trigger and having migraine with aura $\endgroup$
    – Narjes Ali
    Commented Dec 20, 2019 at 20:34

1 Answer 1

Reset to default
1
$\begingroup$

In this case, Pearson's chi-square test is:

A test of independence assess[ing] whether observations consisting of measures on two variables, expressed in a contingency table, are independent of each other.

If this table showed independence, then for "when you eat hotdog" there would be 90.5% of individuals claiming "no relationship" and 9.5% claiming "there is a relationship" whether the individual had been diagnosed or not. The significant chi-square test result indicates that the table does not show independence. For example, 12.9% of those diagnosed claimed "there is a relationship" while only 8.5% of those not diagnosed made that claim. With the size of your study, the chi-square test determines that this difference is statistically significant at p < 0.05.

Clearly, regardless of diagnosis, more people claim "no relationship" than "there is a relationship." But that's not what the chi-square test is examining. Based on your data, it seems that someone diagnosed is more likely than someone not diagnosed to claim "there is a relationship." Nevertheless, the fraction of those in either group who make that claim is still small.

Improve this answer
$\endgroup$
1
  • $\begingroup$ thank you so much!! your answer is so helpful and informative $\endgroup$
    – Narjes Ali
    Commented Dec 20, 2019 at 21:42

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.