A chain of sports centres wishes to estimate the proportion of its 24 000 members who are satisfied with weekend opening hours. Membership records contain a centre identifier and a unique member number. The chain proposes two methods.
Method A: ask the first 200 members who enter the largest centre on Saturday morning.
Method B: select 200 different member numbers using a random number generator from the complete membership database.
Compare the likely representativeness of the two methods.
State one practical limitation of Method B.
Explain why even Method B may produce a result different from the true population proportion.