1: Measures / Variables

What data has been collected?

What summary statistics have been reported?

The data collected in a survey could be of several types, depending on the questions asked.

We will not always be told what the exact questions were, but we can tell what kind of data was collected by the summary statistics that are given in the report.

Sometimes, particular questions or options provided will be quoted in the report, which give further insight into the measures used.

Quantitative Variables

These variables measure a numerical quantity about the respondent. It is either discrete or continuous data.

Qualitative Variables

These variables do not give a numerical quantity as the response. The responses could be sorted, unsorted or open.

Discrete Quantitative

Typically a count, giving a non-negative integer.

For instance, "How many days in the last week did you use public transport?" would give discrete, quantitative data.

Sometimes a Yes/No question can be improved to give more detail. For instance, the question above is an improvement on "Did you use public transport in the last week?"

Continuous Quantitative

A measurement that could be given with arbitrary precision. For instance, the height and weight of a respondent.

Relying on a respondent to take a continuous measurement like height or weight accurately could lead to errors.

Categorical

Sometimes quantitative data is collected with qualitative responses. Broad bands of values are collected into categories.

For instance, this could be ranges of BMI values, rather than the values themselves.

Sorted Response

The respondent is given a list of options to choose from, and the list is sorted in some way. This could be a simple Yes/No question.

For instance, it could be a scale of responses like:

strongly agree
somewhat agree
neutral
somewhat disagree
strongly disagree

Unsorted Response

The respondent is given a list of options to choose from, but the order of the list is not important.

Sometimes the top options on the list may be chosen more often simply because they are more visible, especially on a long list.

Open Response

The respondent is asked a question, but not given any suggested responses.

This can cause difficulties for data entry, in deciding on how to group similar but different responses.

Kiwi shoppers play it cool, survey finds

23 Oct 2014, Natalie Akoorie, NZ Herald (article link)

We’re more wary than the Aussies in giving details to stores ... unless there’s a discount.

Kiwis are careful when handing over personal information to retailers, but are more trusting than Aussies in signing up for a discount.

A survey of 1000 Australian and 500 New Zealand shoppers shows similar expectations of consumers in both countries, except when on personal details.

74% of Kiwis and 68% of Australians rate discount offers as the top reason for providing personal information to retailers.

More Kiwis than Aussies will reveal household structure details, ages of their children and hobbies and interests when joining mailing and loyalty programmes.

But only 6 per cent of New Zealanders compared to 11 per cent of Australians will tell retailers their income and 32 per cent compared to 36 per cent of Aussies will give their date of birth.

159 words

Focus Questions

What variables were measured/collected?
How was the data displayed in the report?
What summary statistics were used?

Possible Answers

The respondents have been asked several yes/no questions about the information they would provide to retailers. They have also been asked why they would reveal personal information, with both countries saying "discount offers" most often.

The percentages saying that they would give each type of information is given in the report, with comparisons between New Zealand and Australia. In some cases, the numbers are not provided, only that "more" Kiwis than Aussies will reveal the information, so more detail would be better in this sentence.

The summary statistics given are all percentages, because there was no quantitative data collected.

Worry question

The subheading implies that New Zealanders will give out our personal information for a discount, but it doesn't seem that the questions specifically addressed this. It would be interesting to see questions like "Would you tell a retailer your date of birth for a 10% discount?" to more directly address this question.

Task 2

Answer the Population measures and variables focus questions about the Salt report:

What variables were measured/collected?
How was the data displayed in the report?
What summary statistics were used?

Report-Salt.pdf

Google Sites

Report abuse