Sampling techniques for primary data collection
Shortcut method for investigating the whole population as there is not enough time to, equipment and labour to measure every single item so the sample must be representable and statically valid to the whole population.
1) the larger the sample the more accurate it is at representing the population
2) A sample with the least bias is the most statistically vaild
3) sample size needs to be balanced between obtaining a statistically valid representation, time, labour and equipment available.
- The least bias as there is no subjectivity
- Each member of the population have equal chance of being picked
- Can be obtained by random number generator
- Can be used with large sample populations
- Avoids the unavoidable bias of personally choosing a sample
- Can lead to poor representation
- The people chosen may not give consent to be in the study and therefore the sample becomes biased with only the ones willing being left.
- Evenly/ regularly distributed in a spatial context
- They can be at equal/ regular intervals in temporarl context
- They can be regularly numbered
- More straight forward to random sampling
- Good coverage of whole sample area
- More biasnas not all members have an equal chance of being selected
- An over or under representation of the population may be created
This method is used when the parent population is made up of sub-sets of known size. These sub-sets make up different proportion of the total population and therefore the sampling needs to be stratefied to ensure the results are proportional and respresentive as a whole.
- Can be used with random or systematic sampling, point or line techniques.
- If proportions of the sub-units are known is can generate results which are more representative of the whole population
- It is very flexible and used in many geographical enquiries
- Correlations and comparisons can be made between the subsets
- The proportions of the subunit must be known and accurate if to be worked out properly
- Hard to stratify questionnaire data collection
- Accurate up-to-date population data must be known and it may be hard to identify peoples age or social background effectively
Using secondary data
Data which already exsists. Sources include libraries, annual reports, Census reports, governmental data,
- Economical- saves time and effort of carrying out own research
- Improves overall data collection as it may have data that you yourself cannot collect due to resources, location etc
- Use it to work out what the gaps are and extra needs in our own data
- Helps improve out understanding of the investigation
- Provides comparison for the data collected by the researcher
- Acuracy of data may be unknown
- Data may be outdated
Remote sensed data (satellite imagery)
Can be used for weather forecasting and predicting natural hazards (storms, volcanic eruptions- bulges in a volcano, earthquakes- movement of tectonic plates)
Data bases (census date)
Able to access huge anount of data that has ben collected over a whole city, region, country and compare with other data sets taken from different places or same data set taken at different times.
Good way of storing a lot of data and recording the growth of cities etc
Can be used by companies and governments when makking policies
Cartographic modelling of land use, soil vegetation, land ownership, characterists.
- Bypass the mechanical proccesses of maping
- Higher quality
- can be acessed anywhere with internet
- Expensive- cant be used all round the world
- Needs regular updating
Use of IT
Uses them for databases, graphical and cartographic images
- Easy to make comparisons over time
- Saves time
- Can be converted into graphs for visual representation
- Easy and quicker to use if the data set is large
- Can be expensive
- Need computers and internet
- Not available everwhere