Population, Sampling & Collecting Data (AQA GCSE Statistics: Foundation): Flashcards

The completed formula is:

Because the strata have to account for the whole population once each, so that their proportions add up to 1.

If the groups overlapped, some members could be counted twice and the numbers taken from the strata would not add up to the sample size you wanted.

A laboratory experiment is run in an environment that the researcher controls.

A field experiment is run in the subject's usual environment, but with the researcher still controlling the situation and some of the variables.

A natural experiment is run in the subject's usual environment with the researcher controlling nothing at all.

Test subjects may not behave naturally in an environment that is not their own, so what the experiment measures may not be what they would really do.

A field or natural experiment is more likely to show usual behaviour, but it gives up some or all of that control.

The completed statements are:

Data is reliable if repeated measurements give similar results, and data is valid if it measures what it was intended to measure.

An open question lets the respondent answer anything at all, so every answer can be different.

A closed question offers a fixed set of answers to choose from, so you can count how many people chose each one and summarise the data easily.

It is a leading question: the wording pushes the respondent towards a positive answer before they have given one.

The responses collected will be biased, so the question should be worded neutrally, for example 'What do you think of our new product?'

They do not cover every possibility, so a respondent who never uses the app, or who uses it more than 5 times a day, has nothing to tick.

The options on a closed question must cover every answer a respondent might give, and they must not overlap, so 'never' and 'more than 5 times' are both needed here.

In an interview the response rate is higher, and the interviewer can explain a question while the respondent can explain an answer.

An anonymous questionnaire can be sent to a much larger sample, and people are more likely to answer sensitive questions honestly because there is no interviewer bias.

A pilot survey gives the questionnaire to a small sample of people first.

It reveals problems with the design, such as questions people misunderstand or refuse to answer, so they can be fixed before the questionnaire is used for real.

The same idea is used before an experiment, where it is called a pre-test.

False.

An anomalous value may be a genuine unusual value that has to be kept, such as one very high salary in a list of company salaries that belongs to the chief executive.

Only values judged to be mistakes should be removed, and any removal has to be justified.

Because a spreadsheet or statistics package reads '12 kg' as text rather than as a number, so it cannot calculate with it.

Cleaning the data first puts every value into a consistent numerical format, and all the final calculations should then be done on the cleaned data set.

Because the members who fail to respond may have something in common, so a whole kind of member ends up under-represented.

If questionnaires are sent to every business on a list, the ones that have closed down cannot reply, so struggling businesses go missing from the results.

An extraneous variable is a variable you are not interested in, but which can still affect the results of your experiment.

Extraneous variables should be identified before the experiment starts and then controlled, so that their effect on the data is removed or at least reduced.

Students timed at the end of the day may run more slowly because they are tired, so a difference in times could be caused by tiredness rather than by height.

The time of day should be controlled by timing every student at the same time of day.