What is sampling?
It is about identifying a subset of a population that can be used to represent the entire group as a whole.
The 3 Sampling Techniques are -
- Random Sampling
- Opportunity Sampling
- Volunteer Sampling
A sample of P’s is produced by using a random technique so every member of the target population has an equal chance of being selected.
+ Representative and unbiased because it is an equal chance method.
+ The R has no influence/control over who gets picked.
- May end up with a biased sample (e.g. more boys than girls as the sample is too small).
- Not everybody may be able to take part so the random may not be as random as you would have liked it to be.
- Time consuming and difficult to do if target population is big.
Uses people from target population available at the time.
+ Simple and doesn’t have to be planned.
+ Cheap and not time-consuming.
- Biased as the sample is drawn from a small part of the target population may not include certain types of people.
- Not representative as only restricted to people who are available at that time.
R places an advert, P’s respond to advert and volunteer to take part by contacting the R.
+ Very little time and effort required from R.
+ Access to a variety of p’s which would make the sample more representative and less bias.
- Sample is biased/unrepresentative because P’s who volunteer are likely to be highly motivates and/or have extra time.
Data which includes numbers and statistics.There are 4 types
1) Nominal: Data which is in categories.
2) Ordinal: Data which is ordered in some way e.g. list in order of liking football teams.
3) Interval: Data is measured using units of equal intervals e.g. counting correct answers.
4) Ratio: Here is a true zero point as in most measures of physical quantities.
+ Easier to analyse because it quantifiable and can be summarised easily.
+ For many R’s this approach to investigating behaviour is the right one as it is regarded as the most scientific and limits the amount of interpretation and opinion and is therefore more objective.
+ Can produce neat conclusions as numerical data reduces the variety of possibilities.
- Oversimplifies reality and human experience (statistically significant but humanly insignificant).
Research techniques which produce this data: Structured observations, case studies, content analysis questionnaires/ interviews (closed q's) and experiments.
Quantitative data Analysis
Quantative data Analysis: Any means of representing trends from numerical data. It can be analysed in 3 different ways:
Descriptive statistics: Allow us to reduce the data into a few numbers so others will not have to spend time reading the raw data trying to understand the results. They can’t do everything they do not tell us what you did or whether findings are reliable the results or type or size of relationship is not explained other.
Examples are -
Measures of central tendency: descriptive statistic which provides info about a ‘typical’ response for a data set (averages).
Measures of dispersion: A descriptive statistic that provides information about how spread-out/dispersed a set of data is.
Visual display: i.e. graphs which provide a way of ‘eyeballing’ your data and seeing findings at a glance
Measures of Central Tendency - Median
Median: The middle value in an ordered set of numbers.
+ Less affected than the mean by extreme scores.
- Not as sensitive as the mean as not all values are reflected.
- Can be unrepresentative in a small set of scores or widely varying scores /doesn’t represent all scores.
Measures of Central Tendency - Mean
Mean: Arithmetic average of a group of scores, calculated by adding up all numbers and dividing by the number of numbers.
+ Most representative/powerful measure since it analyses all the data in its calculation.
- Can be misrepresentative if there are any extreme values (so good to use a measure of dispertion)
- May not make sense in the context of the set of numbers e.g. 2.4 children.
- Cannot be used with nominal data.
Measures of Central Tendency - Mode
Mode: most common number –> if 2 modes=bimodal and more that 2 modes=multimodal.
+ Useful when data is in categories i.e. nominal categories.
+ Easy to calculate.
+ Unaffected by occasional extreme scores.
- Not useful of describing data when there are several modes.
- Doesn’t take into account every score.
- Not useful for small data sets.
- Does not always provide a typical score e.g. a small set of numbers when the most frequent number occurs at either end of a set of scores and is far from the central score.
- Sometimes no mode so it’s best used when there are lots of no’s in the sets of data and there’s likely to be lots of tied scores.