In: Statistics and Probability
Public officials for the city of San Francisco, California, have questioned the accuracy of the year 2000 national census. An SRS of 1000 city residents was used to check the percentage of different ethnic groups living in San Francisco. The pertinent data are given in the table below.
Ethnic Origin Census Percent Observed frequency from sample
Asian 30% 290
Black 10% 65
Latino 35% 400
Native American 5% 30
Anglo 15% 160
Other 5% 55
Use the significance level of 0.05 to test the accuracy of the year 2000 census regarding the different ethnic groups.
null hypothesis:Ho: distribution of different ethnic groups living in San Francisco is same as cesus 2000
alternate hypothesis:Ha:distribution of different ethnic groups living in San Francisco is different from cesus 2000
degree of freedom =groups-1=6-1=5
for 5 df and 0.05 level critcal value =11.070
applying chi square goodness of fit test:
observed | Expected | Chi square | |||
category | Probability(p) | Oi | Ei=total*p | R2i=(Oi-Ei)2/Ei | |
Asian | 0.300 | 290.000 | 300.00 | 0.333 | |
Black | 0.100 | 65.000 | 100.00 | 12.250 | |
Latino | 0.350 | 400.000 | 350.00 | 7.143 | |
Native American | 0.050 | 30.000 | 50.00 | 8.00 | |
Anglo | 0.150 | 160.000 | 150.00 | 0.67 | |
Other | 0.050 | 55.000 | 50.00 | 0.50 | |
total | 1.000 | 1000 | 1000 | 28.89 |
here as test statistic 28.89 is higher than critical value we reject null hypothesis
we have sufficient evidence to conclude at 0.05 level that accuracy of the year 2000 census regarding the different ethnic groups is doubtful