Testing of non-parametric statistical hypothesis

Chi-Square Goodness-of-Fit Test

The products explored under the garment purchase were split into and categorized as children’s, women’s, and men’s wear. Due to differing social roles fulfilled by men and women, it was felt that there might be a major difference in the purchases made by men and women.

Women	Children	Men	Total
23	17	50	90

import scipy.stats as stats


observed = [23, 17, 50]
expected = [30, 30, 30]


x = stats.chisquare(f_obs=observed, f_exp=expected)
critical = stats.chi2.ppf(1 - 0.05, df=2)

print("Chi-Square =", x[0])
print("Critical =", critical)
print("p =", x[1])

Chi-Square = 20.6
Critical = 5.991464547107979
p = 3.3633095185718975e-05

Chi-Square Test of Independence

In this case there are two dimensions (gender and buying designer clothes) and the question asked by the chi-square test for association is whether, in this case, gender and buying cheap clothes are independent of each other.

	Men	Women
Buy designer clothes	37	32
Do not buy designer clothes	20	31

import scipy.stats as stats

data  = [[37, 20], [32, 31]]


x = chi2_contingency(data, correction=False)
critical = stats.chi2.ppf(1 - 0.05, df=1)

print("Chi-Square =", x[0])
print("Critical =", critical)
print("p =", x[1])

Chi-Square = 2.440970285902311
Critical = 3.841458820694124
p = 0.11820309038082191