Research Methods Flashcards
1
Q
Why is psychology a science?
A
Because it employs scientific method
2
Q
What is scientific method?
A
A systematic approach for planning, conducting and reporting research which involves collecting empirical evidence.
3
Q
What is empirical evidence?
A
Data collected directly from observation or experimentation
4
Q
Why is scientific method used?
A
It helps to ensure that data collected is accurate, reliable and Th results obtained are valid.
5
Q
What are the steps in psychological research?
A
- Identify a research problem.
- Write a hypothesis.
- Design the method.
- Collect data.
- Analyse Data.
- Interpret Data.
- Report findings.
6
Q
What is an experiment used for? What is an experiment?
A
To test the cause and effect relationships between two or more variables under controlled conditions. In an experiment, a variable is manipulated.
7
Q
What is a variable? How do experiments relate?
A
Any factor that can change in amount or type over time. Every experiment includes at least one independent and one dependent variable.
8
Q
What is an independent variable? An example?
A
It is the variable that is systematically manipulated in order to assess its effect on the participants’ responses. The iv is often the only difference between the control and experimental groups. When testing the effects of caffeine on sleep quality, the IV is caffeine. One group would be given it and the other wouldn’t.
9
Q
What is the dependent variable? An example?
A
It shows the effects of the independent variable. The changes in the DV are caused by the IV. It is the variable that is measured or the results. Eg. When testing the effect of caffeine on sleep quality, the DV is sleep quality.
10
Q
What is the simplest type of experiment often like?
A
It uses one IV (on two levels) and measures one DV. The two IV levels are often the experimental and control groups.
11
Q
What is the experimental condition?
A
The one where the IV is present. An experiment can contain two experimental groups only.
12
Q
What is the control condition?
A
Where the IV is absent. It provides a standard or basis of comparison for the experimental condition.
13
Q
Why do variables need to be operationalised? What variables need to be operationalised?
A
The variables need to be stated in a way that explains how they will be observed or measured. Both variables need to be operationalised.
14
Q
How would a variable be operationalised?
A
When testing the effect of caffeine on sleep quality:
IV: caffeine— 6 cups of strong coffee a day or 30mg of caffeine in 24 hours
DV: sleep quality— sleep quality as measured by the number of hours slept in a night.
15
Q
What is a hypothesis?
A
An educated guess or testable prediction about the relationship between two or more variables.
16
Q
What are the main features of a hypothesis? (8)
A
- It is testable (the variables can be reliably measured, observed and manipulated)
- it is based on logical reasoning, theory or past research findings.
- it is expressed clearly
- it is usually written as a single sentence
- one hypothesis per IV
- it includes the population
- it has a direction
- it includes both the IV and DV (which don’t have to be operationalised)
17
Q
What are participants?
A
People used in an experiment
18
Q
What is a population?
A
The larger group from which the sample is drawn. In scientific research the population does not refer to all the people in a country or the world but a particular group that has one or more characteristics in common.
19
Q
What is the sample?
A
The subset of the larger group (population) chosen for the study. It is always smaller than the population.
20
Q
How are participants selected for a study?
A
Using sampling or selection.
21
Q
What should a sample be? How should this be achieved?
A
It should be representative of the population. Relevant participant characteristics of the population that may influence the study must be reflected. In order to achieve this random sampling must be used.
22
Q
What is random sampling? How can it be done? Why is it done?
A
It occurs when every member of the population has an equal chance of being selected as a participant in the study (and the selection of one participant doesn’t affect the selection of another). A random number generator or table of random numbers or some other lottery type process is used to do this. It ensures that the sample is representative of the population. If a sample isn’t random, it’s biased.
23
Q
What are advantages of random sampling?
A
- The results obtained are likely to be representative of the population.
- the results can be generalised to the population.
24
Q
What are disadvantages of random sampling?
A
- it is not always necessary or desirable to use it
- time consuming and can be expensive
- large sample is necessary
25
What is convenience sampling or opportunistic sampling?
In some studies it is not convenient, suitable or possible to obtain a representative sample (obtained through random sampling). A convenience sample involves selecting participants who are readily available whiteout employing random sampling.
26
What are advantages of convenience sampling?
* widely used, particularly in pilot research to gain a preliminary indication of responses before conducting the actual study
* quick, easy and inexpensive
* often considered to be adequate when investigating mental processes that are assumed to be similar in 'normal' people
27
What are disadvantages of convenience sampling?
* Most times convenience sampling produces a biased sample, because they only use participants available at a particular time or location.
* a convenience sample is not representative of the target population
* the data may be misleading and the results cannot be generalised to the entire population
* has low external validity
28
When does Stratified Sampling occur?
When the population is divided into distinct groups, or “strata” and then Samples of proportionate size are drawn from within each strata.
29
What is random stratified sampling?
The sample is drawn randomly from the groups. This is most accurate.
30
What is stratified sampling?
The sample is not drawn randomly from groups. That is, the known groups in a population are represented in the same proportions in the sample.
31
What are examples of stratified sampling? (2)
* the sample is truly representative of the population
| * the sample is unbiased
32
What are disadvantages of stratified sampling? (3)
* Very time consuming
* difficult to achieve
* not often used
33
What is participant allocation? What’s important when allocating?
The allocating of participants of the sample into different groups involved in a research study. To keep both groups as similar as possible to ensure nothing else affects the DV.
34
What is random allocation? Why is it used?
This means that every participant in the experiment has an equal chance of being in any of the groups used. It is an important means of experimental control. The participants are allocated to the experimental and control groups and the purpose is that the groups are as similar as possible.
35
What are the experimental designs? (3)
* Repeated measures design
* matched-participants design
* independent groups design
36
What is the repeated measures design?
Each participant is in the control and experimental groups. They are tested at least twice.
37
What are advantages of the repeated measures design? (1)
• EVs and potential CVs are minimised because the same participant is being used in both conditions
38
What are disadvantages of the repeated measures design? (5)
* order effect occurs and counter balancing is required to balance out the order effect
* participant demand characteristics may not be kept constant eg. Guessing the purpose of the experiment and behaving accordingly
* participant attrition occurs
* time consuming per participant and many drop out before completed
* can’t always use this method, depending on the nature of the IV
39
How does matched-participants design work? Who is best suited to this kind of experiment?
Pairs of participants, similar in key characteristics that can influence the IV, are randomly allocated, one to the control and one to the experimental. Often pre-testing is done to determine characteristics that need to be matched. Identical twins are great subjects.
40
What are advantages of matched-participants design? (4)
* participants in the experimental group and control group are very similar on the key characteristics of the study
* twins are very similar
* participant attrition is less common
* there isn’t the need to spread out the time between conditions
41
What are disadvantages of the matched-participants design? (5)
* Even though the pair is matched on the key characteristic, it doesn’t match on all human characteristics so the groups will differ and therefore it may be difficult to compare results.
* pre-testing needs to occur, which is time consuming
* often there is a small sample
* if attrition occurs, then it is a loss of two sets of data, not just one
* not often used, as random allocation is usually enough to ensure that different groups are similar in characteristics
42
What is the independent groups design?
Participants are randomly allocated to the control and experimental groups. Used when undertaking cross-sections, studies (drawing conclusions about a population in one moment of time). A very common experimental research design.
43
Advantages of the independent groups design. (3)
* no order effects
* participant attrition is less common
* not the. Red to spread out the time period between different conditions.
44
Disadvantages of independent groups design. (2)
* less control over participant characteristics
| * large sample is needed
45
What is a cross sectional study? (Compared to longitudinal studies)
Data is collected at one time for all participants and conclusions are drawn about a population in one moment of time. It used independent groups design. It is less time consuming and has a lower attrition rate. A large sample size is needed.
46
What is a longitudinal study? (Compared to cross-sectional)
Data is collected at multiple times for all participants (ie at various ages). Drawing conclusions about a population over time. The repeated measures design is used. Participants are compared to themselves overtime-few participant differences. It is time consuming, high attrition rate and expensive.
47
What is a correlational study?
Studies that identify and describe the relationship between two variables. They do not test cause and effect relationships or manipulate variables. The strength and direction between two variables can be identified.
48
What is a correlation?
A statistical measure of how much two variables are related.
49
What are the directions in a correlational study?
Positive and negative correlations
50
What is a positive correlation?
Two variables change in the same direction
51
What is the strength of a correlation?
Correlation coefficient. A decimal number to indicate strength of relationship. From -1 to +1. The higher the number, the stronger the relationship. The - sign indicates a negative correlation and the + sign indicates a positive relationship.
52
What is an example of a positive and then negative correlation?
More exercise and then greater fitness
Less sleep, more emotions
53
What are extraneous variables? How do they affect experiments?
Variables other than the IV that can cause a change in the DV and therefore affect the results of an experiment. In order for a researcher to conclude that the manipulation of the IV solely causes the change in the DV, all other variables must be controlled or at least minimised.
54
What are EVs? What can they be? What are they not?
They are random and are not caused by a fault in the experimental design. They are often to do with participants in the experiment, such as mood, fatigue, health etc.
55
What are confounding variables? What do they do?
Variables other than the IV that cannot be controlled and have had an unwanted affect on the DV. The effects of the confounding variable are often mixed up with the IV. It is impossible to tell which changed the DV. They are systematic and occur throughout an experiment. They are often related to the design of the experiment.
56
What are some extraneous variables? (7)
Individual participant differences, demand characteristics, placebo effect, experimenter effect, order effect, artificiality, use of non-standardised instructions and procedures.
57
What are individual participant differences? How is it minimised? (4)
Eg. Sex, age, mood, diet, personality, culture, ethnicity, education, motivation. All of these and many more, affect how an individual responds in an experiment. Researchers try to ensure that the influence of these is controlled or minimised before the experiment is conducted by using repeated measures, pre-testing, matched pairs, large sample size.
58
What are demand characteristics? How is this controlled? (2)
A cue or trigger that communicates an expectation of an experiment. Participants behave in a way they feel the situation demands them to based on how they are treated by the researcher or some aspect of the research study. Participants don’t always respond to demand characteristics intentionally or consciously. They behave differently to the way they normally would. Deception, naturalistic observation, placebo
59
What is the placebo effect? What are fake treatments? How is it controlled? (1)
It is an improvement in health or wellbeing die to the participant’s belief that they are receiving an effective treatment- even if the treatment is fake. In an experiment this refers to a change in the responses of the participant due to their belief that they are receiving some kind of experimental treatment. Fake treatments include sugar pills or injections of saline, which have no medical benefit. Single-blind procedure (participant doesn’t know what group they are in)
60
What is a placebo?
A fake drug/inactive substance
61
What is the placebo effect?
The change in a person’s health/responses due to their expectations or beliefs.
62
What is the experimenter effect? How is it controlled? (1)
Researchers expectations can unintentionally influence the way they treat participants and interpret data. This effect is defined as a change in a participants response due to the experimenter's expectations, biases or behaviour rather than due to the IV. The experimenter May unintentionally encourage them. Double blind procedure. Neither participant or experimenter knows which group they are in.
63
What is the order effect? How is it controlled? (1)
In repeated measures experiments, participants are exposed to more than one treatment condition. They may be exposed to the same task more than once under different conditions. The tendency for them to preform either better or worse the second time they are tested. Learning from one condition to the next (practice effect), boredom, lack of motivation, fatigue, carry over effect. Counterbalancing.
64
What is the practice effect?
Learning from one condition to another
65
What is the carry over effect?
The influence of one condition on the performance of a subsequent condition.
66
What is artificiality? How can it be controlled? (2)
Experiments are often conducted in labs in universities. This ensures a controlled environment with minimal extraneous variables in a situation where the participant can be closely monitored however the environment can produce artificial or unwanted behaviours or responses that may not occur in real-life settings (can produce demand characteristics). Artificial environments have low ecological and external validity. Natural observations but can’t control variables as well. Make tasks as natural as possible- make pretend natural setting
67
What is use of non-standardised instructions and procedures? How is it controlled? (1)
When instructions or procedures are non-standardised, they are not the same for all participants (except to the exposure of the IV in the experimental group). Even small variations in procedures can affect participants responses. Both groups must be treated the same way.
68
Why is controlling extraneous and confounding variables important?
It is important to control, anticipate or minimise extraneous and confounding variables in order to measure what happens when the IV is implemented. The more researchers control, anticipate or minimise, the more likely the results obtained are valid and reliable.
69
How can experimental design help minimise EVs and CVs?
One of the most common sources of EVs and CVs is participant differences. One way of controlling this is to ensure that all groups are as similar as possible. This is done through selection and allocation. Eg. Choosing repeated-measures.
70
How can counterbalancing help control the order effect?
Counterbalancing involved systematically changing the order of the treatments or tasks for participants in a balanced way. This counteracts the unwanted effects on performance, caused by the order of the task. There are two types, between participants and within participants. At one time half are in the control and the other in experimental. At second time this switches. This cancels learning.
71
How can standardised instructions be given? What must be kept the same?
Using an automation. Everything must be kept the same, the instructions, the procedure, the techniques of observing and measuring.
72
What are the types of data? (6)
Qualitative data, quantitative data, primary data, secondary data, continuous data, non-continuous data
73
What is qualitative data? What is a disadvantage? Examples?
Describes qualities or characteristics of the data. Is in the form of descriptions, words, meanings, pictures or text (not numerical). Eg. Interviews. It is hard to group or quantify
74
What is quantitative data? What is an advantage? Examples?
Information about the quantities or amounts of what is being studied. It is usually expressed in numbers or units of measurement. Eg. Percentages, means, p values, raw scored. It is easy to group and summarise.
75
What is primary data?
Own findings
76
What is secondary data?
Someone else’s findings
77
What is continuous data? What is used to graph it? What is this?
Weight, height, years, time, speed. Things that flow into one another. A histogram (no gaps and number ranges on the x axis)
78
What is non-continuous data? What is used to present it?
Categorical. Nationality, eye colour etc. decried categories. Uses a bar chart (gaps and categories on the x axis)
79
What is a case study? Who uses them? When are they used? What is an example?
Collecting data on a small number of participants. It is an intense and in-depth investigation of behaviour or events. Case studies are used by mental health professionals to develop a detailed profile of a client and their issues. Often used when large numbers of participants are not available. Eg. Phineas Gage
80
What are advantages of case studies? (6)
- useful way of obtaining detailed and valuable information, particularly with rare conditions
- no manipulation or control of variables
- avoid artificiality and provides a snap shot of real life situations.
- provide insights of how others think, feel and behave in similar situations
- provides information to provide a hypothesis for future research
- good for investigating brain damage.
81
What are disadvantages of case studies? (6)
- cannot be replicated to test reliability
- can’t be used to test a hypothesis
- the process of analysing, summarising and interpreting the obtained info can be time consuming and difficult
- may not reflect typical ways of thinking, feeling and behaving.
- generalising to population cannot be done with certainty
- more susceptible to biased information from the participant or researcher
82
What are observational studies? When is it used? What is the most common type of this? How is this done and why?
Collection of data by carefully watching and recording behaviour as it occurs. It is used when the behaviour in question is easy to observe and record. A common type of observational study is a naturalistic observation. This is where a naturally occurring behaviour is obtained in an unnoticed way. This ensures that the observer doesn’t influence the behaviour being watched. Sometimes observers conceal their presence by just blending in with the crowd.
83
What are advantages of observational studies and naturalistic observation? (6)
- through unobstructive observation, researchers can watch and record behaviours in their natural settings
- people are unaffected by the observer or demand characteristics
- allows researchers to gain more accurate information immediately and overtime
- some behaviours can only be observed naturally and cannot be realistically replicated in a lab
- doesn’t require the cooperation of participants being observed
- low artificiality
84
What are disadvantages of observational studies and naturalistic observation? (3)
- in informed consent is not obtained, privacy is violated
- difficult to determine the cause of behaviour be aide there are so many factors (such as observer bias) that may influence behaviour
- very subjective- low reliability
85
What are self reports? What are examples? What types of questions are used?
Examples are questionnaires, surveys and interviews. Typically questionnaires are structured questions on a particular topic. In interviews, a trained interviewer asks either structured questions (yes/no, close answered questions, similar for all participants) or unstructured questions (free open ended, different for everyone).
86
What are advantages of self reports? (5)
- effective ways of collecting information about how people think feel and behave
- quantitative data is easier to interpret but may not give a true representation of a person’s thoughts of feelings
- questionnaires are objective data
- can collect data from a large number of participants in a short amount of time
- participants can show answers at their own pace and without direct supervision.
87
What are disadvantages of self reports? (4)
- people give socially desirable answers
- people typically need good literacy and/or language skills
- qualitative data is more detailed but is difficult to interpret
- observations are subjective and may be subjected to observer bias.
88
What is reliability?
How consistent a measurement is overtime. how reliable = how consistent
89
What is internal reliability? Example?
The extent of which all items in an instrument (ie questionnaire) contribute equally to the final score. Eg. If all the odd numbered questions had a high correlation with all the even numbered questions then the questionnaire would have high internal reliability.
90
What is inter-rater reliability? Example?
The same result is obtained from a test by anyone who conducts it. Eg. If a person is diagnosed with OCD by one person and schizophrenia by another the test has low inter rater reliability.
91
What is parallel form reliability? Example?
Some tests have more than one form that measures the same characteristic. Eg. If repeated measures is used and a participant is administered the same test at two different times, the order effect is likely to occur. If the participant is administered a similar, but different, test, then the test has high parallel for reliability and the order effect is less likely to occur.
92
What is test-retest reliability?
When a rest is readministered to the same person again at a different time and the result is the same, the test has high test-retest reliability
93
What is validity?
Means accuracy- measures what it is supposed to measure
94
What is internal validity and what are the two types?
Are the results obtained from a measure actually due to the variable that is measured? Eg. Does an IQ test actually measure intelligence.
a) content validity
b) construct validity
95
What is content validity?
Does the instrument measure what it is supposed to measure?
96
What is construct validity?
Can the instrument be used to support theory?
97
What is external validity?
To what extent do the results from this instrument compare with other established instruments.
98
After research has been conducted, what three things must occur?
- the results are summarised, organised and described using descriptive statistics
- results are interpreted so they can be understood using inferential statistics
- results are explained in terms of theory and last experience. Conclusions and generalisations are made.
