Question

9. An internet provider wants to estimate the proportion of its existing customers that would sign up to their new streaming service for movies, tv shows etc. Several sampling strategies have been proposed.

An internet provider wants to estimate the proportion of its existing

For each of these proposed sampling strategies, indicate what type of sampling method is used and what biases, if any, may result.

a) Randomly select one street in each city which has many users of this internet provider and door knock households with this provider asking if they would purchase the streaming.

b) Go through the internet provider’s customer records, selecting every 100th customer and send an email asking if they are interested in purchasing this streaming.

10. A supermarket has a loyalty card for their customers who shop in person or shop online. The supermarket wants to survey their loyalty card customers about their online shopping service. Specifically, the supermarket wants to know what proportion of loyalty card customers plan to use their new online order/delivery system. The supermarket takes a random sample of 800 customers from the loyalty card database and sends an email with a request to fill out a survey in exchange for a $50 online supermarket voucher.

a) What is the population of interest?

b) What is the sampling frame?

c) What sampling method is used?

d) Will this sample suffer from any bias? Explain.

11.A student is trying to find the average mark of a test in BUS 108. They ask 3 students their test mark and find the average mark. The student now decides to ask 10 students their test mark and finds the average mark. Why do you think the student decided to increase the sample size?

12. Decide which of the following is cross-sectional data or time series data.

a) $75 AU, the value of the Australian dollar today.

b) Monthly Wine sales (millions litres) for the same company in 2018: c) $650, your last electricity bill for your home

d) Your electricity bill ($) for each quarter of 2018

13. An online site published an article comparing smart phones. It selected 40 of the most popular smart phones currently for sale, comparing the brand, name, cost (in dollars), size of screen (in centimetres), camera resolution (in pixels) and colours available, amongst other things.

a) What are the categorical variables? Are they nominal or ordinal?

b) What are the numerical variables and their units of measurement?

c) State whether the 40 mobile phones are a sample or population. Briefly

explain your answer.

14. Classify each of the following as nominal, ordinal, discrete or continuous data.

a) Number of laptops in a household

b) Price of Apple Inc. shares

c) Student ID number

d) Assembly time of a computer chip

e) Postcodes of different suburbs

f) Restaurant reviews.

15. Which of the following would you consider as primary data or as secondary data?

a) Data on employees recorded by the Human Resources department of a large corporation, used by payroll.

b) Results from a survey on washing powder conducted by the same washing powder manufacturing company.

c) Share prices for an Australian bank over the past seven years, posted on a website.

d) Voting history of different electorates in a city.

Answer 1

Multiple Qs. Solving 1st one.

9.

a) Randomly select one street in each city which has many users of this internet provider and door knock households with this provider asking

Sampling method: Clustered sampling because here subgroups of the population are used as the sampling unit. Cluster sampling is prone to biases: Streets are randomly selected based on high users which is formed under a biased opinion.

b) Go through the internet provider’s customer records, selecting every 100th customer and send an email asking if they are interested in purchasing this streaming.

Sampling method: Systematic sampling- Individuals are selected at regular intervals from the population.

It may also lead to bias: if there are patterns in the order of the users in the sampling frame, such that the sampling technique coincides with the periodicity of the underlying pattern. Example: in case there is periodic pattern in the way customer list is organised in company's database then picking every 100th user may result in selecting more from a particular user group.