3.2 Collecting primary data Flashcards
Sampling
- process of selecting a portion of a population being studied, to make generalisations of the population
- always a subgroup of the population
- necessary cause it is not possible & practical to obtain info from the entire population
Sampling methods
- Probability sampling
1. Simple Random Sampling
2. Stratified Random Sampling - Non-probability sampling
1. Convenience Sampling
2. Quota Sampling
Probability Sampling
- samples are randomly selected, w/o researcher’s conscious decision
- done using random no generator or dice
- removes bias
- greater chance of creating a representative sample
- used when you want a more representative sample
Non-probability sample
- non-randomly selected, using researcher’s conscious decision
- researcher may subjectively select samples
- selection may be biased
- unlikely to be representative since they are bias, making it hard to make generalisations about the population
- used when it is impractical/ unnecessary to select a representative sample
- used when sampling needs to be conducted quickly due to time limitations.
Simple Random generator
- give every member of population a number
- random number generator used to generate random numbers to select samples
Stratified random sampling
- select sample that is proportional to population based on age and sex (if 65% of population is male, 65% of sample is male)
- if random sampling is not used, it becomes quota sampling
Quota sampling
- select sample proportional to population based on age or sex or any other category
- uses convenience sampling to select sample instead of random sampling
Convenience sampling
- samples are selected cause they are convenient sources of data
- like friends and people walking down a street
Questionnaire surveys
- investigate opinions of people and organisations using a series of question
- used to collect data with questionnaire surveys using:
1. predefined responses
2. rating scales
Predefined responses
- could be short-phrases, or a single word arranged in a series, or numbers
- actual values can also be recorder when residents are asked to answer questions with a number
Advantages:
- these questions guides participants, making it easier for them to answer
- easier to analyse and interpret responses since they are put into fixed categories
- useful for quantitative data analysis to examine patterns and trends
Rating scales
- a set of pre-defined responses
- used to guide survey participants to respond to questions on their opinions with a wide range of responses
- Frequency scale: present a range of options based on no. of occurrences
- Ranking scale: get people to compare items w/ one another and rank them (ranking scales are better to be kept small, less than 10 items, to produce reliable data)
- Likert Scale: people will state how satisfied they are/ how much they agree with a statement
Mental maps
Used to collect data on:
- how peeps experience and think visually and spatially about their environment
- dynamic interrelationships between people and their environment
Participants are given:
- blank paper and asked to draw features in a map form like how they know, believe and/or feel about a place/places
- a base map and asked to add details by labelling or annotating their perceptions of places on map (easier & less time consuming)
- depending on research question/ hypothesis, peeps are given diff base maps, instructions and tools to create metal map
- free-form mental maps are more representative of their perception of places then those drawn w/ base maps
- however, free-form mental maps are not easily geo referenced and mapped onto Geographical Info System
- semi-structured interviews w/ open-ended questions can be conducted after to find out more about peep’s perception of the places
