Sorry I have no statistical analysis expertise. My work is in practical learning and teaching, supported by critical and participative observation and collaboration: see ‘Inside Teaching: How to Make a Difference for Every Learner and Teacher’ (Routledge, 2017).

If I understand your intention correctly, you are testing the association of the responses to the two items.

For simplicity, assume that the answer to #1 is either “Has outdoor activities” or "Has NO outdoor activities". And, assume that the answer to #2 is either "Address child's needs" or “Dose not address child's needs". Then you have four possible situations.

With these, for a 2-by-2 contingency table for the chi-square test, the NULL hypothesis could be "There is NO association between outdoor activities and addressing child's need." Of course, you hope your data does not support the Null so that you may conclude that outdoor activities are associated with address of child's needs.

For a 2-by-2 chi-square, you can estimated the phi-coefficient which is an estimate of the Pearson's correlation (r).

Of course, you may have more than two categories for the answers, then you have a r*k contingency tables of higher dimensions for the chi-square test.

By the way, this is the chi-square test of association, not of goodness of fit; the concepts are different although the obtained statistics may be the same.

Hope this helps.

Interesting topic.

Sorry, I focused in science education and supported by research in teaching, learning, and professional development in education

Very interesting topic

Hi, I agree with Kay.You should clearly identify your independent and dependent variables.

You need to consider reviewing literature then from the reviewed literature. You may now have the hypothesis.

Searching for independence testing, like nominal association.

Hi, agrees with Kay's answer.

We must to identify towards on position between two variables in your research

In methodology, I agree with Kay's reply. But whether the outdoor activities are associated with addressing children's needs should be grounded on the relevant research literature.

Poorva，

The result of chi-square test only tells whether the two variables (Outdoor activities, children's needs) are associated. Which of the two is assumed to be the cause and the other the effect? Do you have reasons to make the assumption? As Donghai suggested, you need to go to the pertinent literature to sort this out.

Yes. Kay. Some of the calculation results would be statistically sound but conceptually irrelevant.

With statistics software so readily available, calculation is not a problem. It is the correct interpretation of the resultant statistics that counts.

Often the early problems are the types of data, which come to the accurate conclusion. To formulate the hypothesis means the research or the design is logically valid.

It is better for closed questioner. So, the instruments need small measurable indicators. You can use non-parametric statistics analysis, manually or computer package.

The questionnaire is often related to descriptive research, and descriptive research often does not have hypotheses.

Good answer Dr. Mohamad Rif'at .