Sampling methods and research methods (P3)

Naturalistic Observation.

What is it: Participants observed in their natural environment while carrying out behaviour, may be overt or covert. Usually targets specific behaviour or set of behaviours, researchers often use tally chart or grid to measure how many times spefici behaviour is carried out.

Advantages: Behaviour is naturally occurring and is not manipulated by a researcher and it can provide more qualitative data as opposed to merely quantitative information.

Limitations: Even the presence of someone observing can cause those being observed to alter their behaviour. Researcher's beliefs can also alter their observations. And, it is very difficult to coordinate multiple observers since observed behaviours must be operationally defined (e.g. what constitutes an aggressive act)

Laboratary Experiments

Researchers changing an independent variable to measure the change in a dependent variable, produces a cause and effect relationship.

Advantages - Tighter control of variables. Easier to comment on cause and effect. Relatively easy to replicate.

Enable use of complex equipment.

Often cheaper and less time-consuming than other methods.

Limitations- Demand characteristics - participants aware of experiment, may change behaviour. Artificial environment - low realism. May have low ecological validity - difficult to generalise to other situations. Experimenter effects - bias when experimenter's expectations affect behaviour.

Field Experiments

Still and IV that is manipulated to cause change in DV but takes place in natural setting of behaviour, such as a school

Advantages- People may behave more naturally than in laboratory - higher realism. Easier to generalise from results.

Limitations - Often only weak control of extraneous variables - difficult to replicate. Can be time-consuming and costly.

Natural experiments

No manipulated IV, rather it exists in sample: like taxi driver or non taxi driver, eating disorder rates before and after event
Advantages- Situations in which it would be ethically unacceptable to manipulate the independent variable. Less chance of demand characteristics or experimenter bias interfering.

Limitations- The independent variable is not controlled by the experimenter. No control over the allocation of participants to groups (random in a 'true experiment').

Case Study

Research into specific person, group or organisation, more detailed and holistic overview of one specific instance, often using a variety of methods
Advantages: Can gather extensive information, both qualitative and quantitative and it can be helpful in better understanding rare cases or very specific interventions

Limitations: Only one case is involved, severely limiting the generalization to the rest of the population. Can be very time consuming and can involve other problems specific to the techniques used, including researcher bias.

Survey

Set questions that the participant answers either online or by text, so no interaction with researcher

Advantages: Can gather large amounts of information in a relatively short time, especially now with many surveys being conducted on the internet.

Limitations: Survey data is based solely on subjects' responses which can be inaccurate due to outright lying, misunderstanding of the question, placebo effect, and even the manner in which the question is asked

Correlational Studies

Test the relationship between two variables through mathematics and existing data

Advantages: Can assess the strength of a relationship. Is popular with lay population because it is relatively easy to explain and understand.

Limitations: Can not make any assumptions of cause and effect (explain how third a variable can be involved, or how the variables can influence each other).

Interviews (SS and US)

SS: Both open-ended and close-ended questions, open-ended allow for elaboration and rich data whereas close ended more concise
US: General topics to discuss but more like a convo where researcher just guides it

Advantages- Detailed information can be obtained and avoids oversimplifying complex issues.

Greater attention to individual's point of view this is important in clinical psychology.

Unstructured, casual interviews may encourage openness in answers, mimicks conversation so ecologically valid data, good if rapport is established

Disadvantage- Difficult to analyse if unstructured and qualitative in nature.

Time-consuming, expensive.

Possible interviewer effects. For example, people affected by attractiveness of interviewer!

Questionnaires

Advantages- Many people can be tested quickly. It is easy to generate quantitative data and easy to analyse.

Used to collect large amounts of data about what people think as well as what they do!

Convenient - researcher does not need to be present as answers can be mailed so respondent has time to consider answers.

Can quickly show changes in attitudes or behaviour before and after specific

Disadvantages- Social desirability - people say what they think looks good.

People may not tell the truth, especially on sensitive issues, for example, sexual

behaviour.

If researcher is present then this may affect answers. Also, postal surveys may have low response rate.

Difficult to phrase questions clearly, you may obtain different interpretations of questions

Random Sampling

a sample in which everyone in the target population has the same probability of being chosen

PRO: best used in quantitative research, limits sample bias, makes sure sample is representative of target population

CON: Not really possible if target population is too large, participants may refuse to take part, Very difficult to obtain a truly random sample, as researchers may not have details of ALL the people in their target population

Stratified Sampling

sampling technique that attempts to make a sample that reflects the sub-groups within a target population

PRO:sample would be highly representative of the target population and results can be generalized

CON:gathering this sample can be difficult and time-consuming

Opportunity (Convenience) Sampling

Using a group of participants not randomly selected from the target population, but invited to participate because they are easily contactable

PRO:quick and easy

CON:groupings tend to be relatively homogeneous and not usually representative of target population, participants may want to please the investigator who may be known to them.

Purposive Sampling

individuals who are expected to offer the most detailed or most appropriate information will be approached and invited to participate. used when researchers have a clear idea of who they want to study

PRO:sometimes the most effective way of obtaining research data, ensures that all participants are salient with specific criteria, using the participants who are most relevant to the research question can increase the validity

and reliability of the results.

CON:subject to researcher bias (but can be okay if the technique has been acknowledged and justified by researcher)

Snowball Sampling

type of purposive sampling where participants are asked to invite other people they know to participate, recruited through the researcher’s own contacts and participants’ social networks

PRO:helps researcher access participants that may not normally volunteer; helps build trust with researcher, helps build salient sample quickly

CON:bias (can be acknowledged and justified)

Self-selected (volunteer) Sampling

sampling technique that asks people to voluntarily

PRO:people that sign-up are usually motivated and less likely to drop out of the research study

CON:sample may not be representative of target population, they be more obedient or more motivated to take part in studies

Cluster Sampling

randomly selecting one section of the target population and inviting members of that cluster to participate
PRO:easier and faster than other sampling methods
CON:cluster selected may be biased and non-representative