Topic 7 - Research methods Flashcards
Define ‘Researchers aim’
Once a researcher has an initial idea about an area of interest, the next step is to narrow the focus of our research to produce an aim.
Define ‘Types of hypothesis’
They are general statements that describe the purpose of an investigation.
Define ‘One-tailed (directional) hypothesis’
Predicts the nature of effect of the independent variable on the dependent variable (faster or slower; more or less; increase or decrease).
Define ‘Two-tailed (non-directional) hypothesis’
Predicts that the independent variable will have an effect on the dependent variable, but the direction of the effect is not specified (unsure of the difference that will be caused). This would be used when there is no previous research or there is contradicting evidence in the area.
Define ‘Operationalisation’
Many of the things’ psychologists are interested in studying are not easy to define. One of the main tasks of researchers is to ensure variable are as measurable as possible.
Define ‘Pilot study’
A small-scale trial run of an experiment to test variables, conditions and methods.
Define ‘Single-blind procedure’
Participants will not be told the aim of the experiment, which condition they are in, or even if there is another condition at all.
Define ‘Double-blind procedure’
Both researchers and participants will not be told the aim of the experiment. Double blinds are an important feature of drug trials in which the drug will be administered by a third party.
What are demand characteristics and investigator effects?
Presence of demand characterises in a study suggest that there is a high risk that participants will change their natural behaviour in line with their interpretation of the aims of a study, in turn affecting how they respond in any tasks they are set. Investigator effects are where a researcher (consciously or unconsciously) acts in a way to support their prediction. This can be a particular problem when observing events that can be interpretated in more than one way.
What is an independent groups design?
An independent groups design uses different participants in each of the experimental conditions, so that each participant only does one condition (either the experimental or control condition). Different participants are therefore being tested against each other.
Give two advantages of independent groups design.
- No order effects
- Time is saved
- No demand characteristics.
Give two disadvantages of independent groups design.
More participants needed & group differences.
What is a repeated measures design?
Repeated measures design each participant is tested in all conditions of an experiment. Participants are therefore being tested against themselves.
Give two advantages of repeated measures design.
Give two disadvantages of repeated measures design.
What is a matched pairs design?
Matched pairs design is a special kind of IGD. Different, but similar, participants are used in each condition. Participants are matched on characteristics important for a particular study, such as age. Identical (monozygotic) twins are often used as they form perfect matched pairs, sharing identical genetic characteristics.
Give three advantages of the matched pairs design.
Give three disadvantages to the matched pairs design.
- More participants needed.
- Matching is difficult.
- Time-consuming.
What is quantitative data?
Quantitative data are scientific and objective. Numeric measures are used, and data can be statistically analysed. Data is high in reliability. However, quantitative data may lack detail and is often collected in artificial settings.
Give advantages of using quantitative data.
- Less open to bias.
- Easy to analyse.
- Quick & usually cheap.
Give a disadvantage of quantitative data.
- May be over simplified.
- Doesn’t provide explanations.
What is qualitative data?
Qualitative data is rich and detailed and often collected in real-life settings. Data is high in validity. It can provide information on people’s attitudes, opinions, and beliefs. However, qualitative data may be subjective and an imprecise measure and may be low in reliability.
Give two advantages to qualitative data.
Rich detailed information & helps researchers to tease out underlying meaning.
Give two disadvantages to qualitative data.
Open to bias & complex to analyse.
What is primary data?
This data collected first hand by the researcher.
Give two advantages of primary data.
Control over it and its easily understandable.
Give a disadvantage of primary data.
Expensive.
What is secondary data?
This data is information that was collected for a purpose other than the current one.
Give three advantages of secondary data.
Quicker, easier, and cheaper.
Give two disadvantages of secondary data.
Outdated, not relevant.
What is stratified sampling?
Means layer. A stratified sample is made up of different layers of the population. For example, selecting people from different age groups or ethnicities so it mirrors the distribution of groups in a larger population.
Give two advantages of stratified sampling.
- Unbiased selection- as random sampling is performed upon the sub-groups of a population, selection is unbiased.
- Generalisation – the sample should be representative, results will be generalisable to the target population.
Give two disadvantages of stratified sampling.
- Time consuming.
- Knowledge of population characterises.
What is opportunist sampling?
Using anyone available.
Give two advantages of opportunist sampling.
- Ease of formation.
- Natural experiments.
Give two disadvantages of opportunist sampling.
- Unrepresentative- likely to be biased.
- Self-selection.
What is volunteer sampling?
Participants volunteer themselves.
Give two advantages of volunteer sampling.
- Ease of formation
- Less chance of ‘screw you’ phenomenon – sabotage the study.
Give two disadvantages of volunteer sampling.
- Unrepresentative – likely to be biased.
- Demand characteristics – volunteer is eager to please.
What is random sampling?
Means that members of a population have an equal chance of being selected. To do this, a random number table/generator is used to select participants.
Give one advantage of random sampling.
Unbiased selection – increasing the chances of getting an unbiased and thus representative sample.
Give one disadvantage of random sampling.
- Impractical, difficult to achieve as it is difficult to get full details of a target population and not all members will be available.
What is systematic sampling?
Involves taking every ‘nth’ item from the sampling frame, e.g. Every tenth name from a register.
Give one advantage of systematic sampling.
- Unbiased selection – increasing the chances of getting an unbiased selection.
Give one disadvantage of systematic sampling.
- Not representative – unbiased selection does not guarantee an unbiased sample.
What is a lab experiment?
An experiment that takes place in a controlled environment within which the researcher manipulates the iv and records the effects on the dv whilst maintaining strict control of extraneous variables.
Give two advantages of lab experiments.
- High degree of control – increases internal validity – what takes place within the study is more accurate.
- Replication – easy for other psychologists to use (simple method, like making a cake). This improves reliability.
- Cause and effect – easier to establish as a result of the controlled setting
- Isolation of variables – makes it very clear and easy to establish the impact of the iv on the dv.
Give two disadvantages of lab experiments.
- Experimenter bias – researcher may become too involved with the study. Influence participants and interpret results subjectively.
- Problems operationalising the iv and dv - if categories are too vague – validity and reliability are negatively impacted.
- Low external (ecological) validity – external validity = extent to which findings can be applied outside of the lab experimental.
- Demand characteristics – pps change their behaviour as a result of being studied.
What is a field experiment?
An experiment that takes place in a natural setting within which the researcher manipulates the iv and records the effects on the dv.
Give three advantages of field experiments.
- Mundane realism – occur in everyday situations and in places where behaviour is most naturally occurring.
Less demand characteristics – If covert (unaware) – behaviour is more natural. Not acting in the presence of a researcher.
External validity – it is likely that the findings are widely generalisable due to the natural setting/task.
Give three disadvantages of field experiments.
- Sample bias – may be opportunistic/volunteer sample and the sample may not be representable
- Loss of control of extraneous variables – cannot control/predict all environment stimuli, validity is decreased as impact of iv and dv becomes less clear.
- Ethical issues – consent may be an issue if the research is covert, sensitive nature of participants personal environments must be considered.
What is a natural experiment?
An experiment in which the change in the iv is not brought about by the researcher but would have happened even if the researcher had not been there. The researcher records the effects on the dv.
Give three advantages of natural experiments.
- Mundane realism – occur in everyday situations and in places where behaviour is most naturally occurring.
External validity – it is likely that the findings are widely generalisable due to the natural setting/task. - Opportunities for research that may not arise elsewhere – natural experiments are often used when it is unethical to manipulate an iv. Therefore, providing insight into these areas.
Give three disadvantages of natural experiments.
- Loss of control of extraneous variables.
- No random allocation to experimental conditions – due to the iv being naturally occurring they’re cannot be any random allocation.
- Replication – since the conditions are never exactly the same again.
What is a quasi-experiment?
A study that is almost an experiment but lacks key ingredients. The iv has not been determined by anyone – the ‘variables’ simply exist, such as being old or young. Strictly speaking this is not an experiment.
Give two advantages of quasi experiments.
- External validity – high ecological validity due to real world environment, results relate to everyday behaviour and can be generalised to other settings.
- Opportunities for research that may not arise elsewhere – often used when it is unethical to manipulate an iv. Therefore, providing insight into these areas.
Give one disadvantage of quasi experiments.
- No random allocation to experimental conditions – due to the iv bring naturally occurring there cannot be any random allocation.
What is internal validity?
Refers to whether the effects observed in an experiment are due to the manipulation of the iv and not some other factor.
For example: Milgram’s results were confounded by participants ‘playing along’ rather than responding to real shocks.
What is external validity?
Relates to factors outside the investigation, such as generalising to other settings, other populations or other eras. For example: a major threat to this is participants responding to demand characteristics.
What is ecological validity?
Refers generalising findings from one setting to another setting, most particularly to ‘everyday life’. For example: asch’s research was confounded by the task being artificial.
What is temporal validity?
Refers to whether the findings from a study hold true over time. For example: freud’s concept of ‘penis envy’ is deemed outdated as it was a product of a patriarchal victorian society in which it was developed.
What is face validity?
Concerns the issue of whether a self-report measure looks like it is measuring what the researcher intended did to measure. For example: if the questions on the questionnaire are obviously related to stress.
What is concurrent validity?
Involves comparing the current method of measuring stress with a previously validated one on the same topic. To do this, participants are given both measures at the same time and then their scores are compared. We would expect people to get similar scores on both measurement thereby confirming concurrent validity of the current questionnaire.
What are extraneous and cofounding variables?
- Extraneous variables are controlled so that they do not vary across any of the experimental conditions or between participants.
- Uncontrolled extraneous variables can become confounding variables and ‘confuse’ the results by affecting the dv.
What are participant variables?
Concern factors such as participants age and intelligence.
What are situational variables?
Concern the experimental setting and surrounding environment, for example the temperature and noise levels.
What is inter-observer reliability?
Measures the degree to which different rates give consistent estimates of the same behaviour.
What is test-retest?
Administering the same test or questionnaire to the same person on different occasions. If the test is reliable then the results will be the same or very close across time and occasions.
What is split-half reliability?
Measures the extent to which all parts of the test contribute equally to what is being measured.
What are unstructured (natural) observations?
Researchers record any behaviour they can see. For example, you could wander into a classroom and capture anything and everything in your notes/recording.
Give two strengths of unstructured (natural) observations.
- Provides a great deal of rich, in depth, qualitative data.
- High external validity.
Give three weaknesses of unstructured (natural) observations.
- Lack of control makes replication difficult (low external validity)
- Extraneous variables – limits ability to measure impact of iv.
- Time sampling - recording behaviour within a particular time frame.
- Event sampling - recording each time certain event happens.
What are structured (controlled) observations?
Continuous recording of behaviour is a key feature of unstructured observation. A researcher can achieve this in two ways.
What is time sampling?
Recording behaviour within in a particular time frame, eg. Every 30 seconds
What is event sampling?
Recording each time a specific behaviour occurs, eg. Smiling
What are covert observations?
Observation that takes place without the participant being aware.
Give two strengths of covert observations.
- No demand characteristics/participant reactivity as participants are unaware.
- High external validity - natural behaviour in a natural environment
Give one weakness of covert observations.
- Ethically questionable - issue of informed consent
What are overt observations?
Participants give their informed consent for their behaviour to be observed and recorded.
Give one strength and one weakness of overt observations.
Strengths:
- Ethically sound
Weaknesses:
- Lacks validity due to demand characteristics - participants are aware they are being watched (social desirability)
What is a non-participant observation?
The researcher remains separate from the group they are studying, as it is not always practical or possible to join the group.
Give one strength and one weakness of non-participant observation.
Strengths:
* Allows the researcher to maintain an objective distance from their participant.
Weaknesses:
* Validity may be low due to demand characteristics.
What is a participant observation?
The researcher joins the group they are studying, in order to gain more first-hand data.
Give one strength and one weakness of participant observation.
Strengths:
* Highly valid - experience the behaviour first hand with no demand characteristics.
Weaknesses:
* Ethically questionable - the researcher may be in too deep and risk ‘going native’.
What is a case study?
Often involves analysis of unusual events or individuals. It usually generates qualitative data by constructing case history (interviews, observation, and questionaries) it may include psychological testing which can produce statistical data. Can take place over a long period of time and include additional data from family and friends.
Give two strengths and two weaknesses of case studies.
Strengths:
* In depth data that can provide new insights.
* Investigating human behaviour and experience that is rare and provides us insight into cases that would be ethically unacceptable to conduct.
* Case studies can generate hypothesis for future study.
Weaknesses:
* Case studies lack generalisability - unique cases.
* Ethical issues with regards to confidentiality and informed consent – unique characterises of individuals.
* Subjectivity and researcher bias.
* Case studies lack reliability – no way to repeat.
What are the key features of content analysis?
- A type of observational research in which people are studied indirectly through mediums they have produced.
- Content analysis can be conducted on a wide range of mediums, some of the most common include songs, films, tv shows
- The key aim of content analysis is to summarise and describe the medium in a systematic way overall conclusions can be drawn.
Give two strengths and two weaknesses of content analysis.
Strengths:
* Ease of application – inexpensive and easy to perform.
* Reliable – easy to replicate, through others using the same material.
Weaknesses:
* Descriptive – does not reveal underlying reasons.
* Flawed results – is limited by availability of material.
What is coding?
- Coding involves categorising information into meaningful units.
- Involves generating qualitative data – thematic analysis.
- Information coded ‘themes may begin to occur.
- A theme refers to an idea or concept that keeps cropping up in the data and re more descriptive than coding units.
What are open questions and give some detail about them.
Allow the participant to explain their answer, generate qualitative data.
* Include greater depth/detail.
* Can be difficult to collate/summarise.
* Conclusions drawn may be bias and open to bias.
What are closed questions and give some detail about them.
Usually yes/no answer. Generate quantitative data.
* Easier to compare.
* Quick to complete.
* Respondents can’t explain answers.
What are the three big issues with designing questions for a questionnaire?
Issue 1: overuse of jargon.
Issue 2: emotive language and leading questions.
Issue 3: double barrelled questions and double negative.
Name the three types of interviews and explain what they are like?
Structured: where the interviewer sticks rigidly to a pre-written set of questions (high reliability and standardized)
Unstructured: without any pre-planned questions. Instead the interviewer might have some headings to guide a free-flowing discussion. (less standardized, make people feel more comfortable)
Semi-structured: where the interviewer does have a list of questions, they intend to ask, but they are happy to ask supplementary questions if required or to leave out questions if deemed necessary.
What is a likert scale?
The one in which respondent indicates their agreement with a statement using a scale of usually five points. The scale ranges from strongly agree to strongly disagree.
What are rating scales?
They work in a similar way but gets respondents to identify a value that represents their strength of feeling about a particular topic.
What is informed consent?
- Means revealing the true aims of studying and/or telling them what is going to happen.
- Researchers often do not reveal all aspects of study to the participant to decrease demand characteristics.
- Should be told what they will be required to do before consenting to participating in the study. Also, should be told the risk involved.
- However, epstein and lasagna (1969) found that only a third of participants volunteering for an experiment really understood what they had agreed to take part in.
What is deception?
- It can be necessary to deceive about the true aims of a study otherwise participants may alter their behaviour.
- However, this rages in how acceptable it is (withholding some of details – reasonably acceptable, deliberately providing false information – less acceptable).
- Deception prevents participants being able to give informed consent. They may agree to participate without knowing what they have really let themselves in for and this could be a potentially distressing experience.
What is the right to withdraw?
- If participants leave during the study this may bias the results
- If those who stayed may be more obedient or hardier. Which will lead to a bias sample.
- If a participants made feel uncomfortable or distressed, they should be able to withdraw.
- This is especially important if the participants have been deceived.
- Some participants who have received money or university credits may feel less able to withdraw.
What is confidentiality?
- It may be difficult to protect confidentiality because the researcher wishes to publish the findings.
- A researcher may guarantee anonymity but even then, it may be obvious in a study.
- The data protection act makes confidentiality a legal right.
- It is only acceptable for personal data to be recorded if the data is not made available in a form that identifies the participants.
What is a positive correlation?
When one variable increases the other variable increases.
What is a negative correlation?
When one variable increases the other decreases.
What is a correlation?
A relationship between two co-variables.
Give two strengths of correlations.
- Useful when it would not be practical or ethical to do an experiment-because the variables cannot be manipulated
- Starting point of research/suggestions for future research
- Quick and cheap to carry out - no need for a controlled environment, can use secondary data
- Allow researchers to make predictions about similar situations.
Give two weaknesses of correlations.
- Correlations cannot show causation.
- People often misinterpret correlations and think if two variables are related one must cause the other; however, we cannot say this because extraneous variables have not been held constant.
What are descriptive statistics?
A way of describing quantitative data and identifying any patterns or trends using graphs, tables and summary statistics.
What are the two ways that descriptive statistics can be measured?
- Measures of central tendency – the general term for any measure of the average value in a set of data.
- Measures of dispersion - based on the spread of scores.
Give one advantage and one disadvantage of the range being used to calculate measures of dispersion?
(+) easy to calculate and provides a margin of error (+1)
(-) is affected by extreme values – outliers.
Give one advantage and one disadvantage of standard deviation being used to calculate measures of dispersion?
(+) a much more sophisticated measure of dispersion because it takes into account all the exact values in a data set.
(-) like the mean, sd can be distorted by extreme values.
What is a negatively skewed (right skewed) distribution?
The bulk of the scores are concentrated on the right, resulting in a long tail of anomalous scores on the left. The mean is pulled to the left (due to the lower scorers who are in the minority), with the mode dissecting the highest peak and the median in the middle.
What is a positively skewed (left skewed) distribution?
This is where most of the distribution is concentrated towards the left of the graph, resulting in a long tail on the right. Imagine a very difficult test in which most people got low marks with only a handful of students at the higher end. The mode remains at the highest point of the peak, the median comes next followed by the mean (this is due to the outliers pulling the mean to the right as it includes all the scores).
Name the 5 types of graphs.
- Bar chart
- Line graph
- Scattergram
- Histogram
- Pie chart
What is the equation for calculating percentage change?
Percentage change = Change × 100
Original
What is a peer review?
The final stage of psychological research. This is the point at which research findings are subject to scrutiny from a panel of specialists.
What is the sign test and when is it used?
- This is used when looking for a difference (e.g. In an experiment and not a correlation)
- It is used in a repeated measures design.
- We need nominal data – data that is in categories.
Explain the steps you would need to take to complete the sign test?
- Step 1: record the data.
- Step 2: convert to nominal data by stating the sign (- or +). If the figure in the right column is higher, put “+”, if the figure is lower, put “-”.
- Step 3: add up how many + you have and how many – you have.
- The smaller number is your calculated value (observed value) and needs to be compared to a critical values table. It is written as: s =
- Step 4: using a critical values table you have to find the critical value. N = the total number of participants. (note: if anyone has the same score (= instead of + or –) then the data is ignored and the n is adjusted down.
- You need to refer to the hypothesis and find out if it is directional (one tailed) or non- directional (two tailed).
- Go down the 0.05 (unless it is a drug trial, then use 0.01!) Column for the correct tailed test and go across at the correct n value.
What are inferential statistics?
Ways of analysing data using statistical tests that allow the researcher to make conclusions about whether a hypothesis was supported by the results.
What is nominal data?
Can also be referred to as categorical data. If the data is nominal, each participant will only appear in one category.
What is ordinal data?
Data that are ranked in some way and the intervals between the data are not equal. For example, ranking people from tallest to shortest. Ordinal data lacks precision because it is subjective rather than using objective measures.
What is interval data?
Also refers to data ordered in some way, however the intervals between each data are equal. For example, measuring people’s height using centimeters. It is based on numerical scales that include units of equal, precisely defined sizes.
Out of the 8 tests, name the three which are parametric (interval) data.
- Pearson’s r
- Related t-test
- Unrelated t-test
Out of the 8 tests, give the five that are non-parametric (ordinal or nominal) data.
- Chi squared
- Sign test
- Spearman’s rho
- Wilcoxon
- Mann whitney
What are parametric tests?
Favoured as they are more powerful - they make calculations based on the mean and standard deviation of a data set.
What are non-parametric tests?
Use ranked data, thus losing some detail. Parametric tests perform a more sensitive and accurate analysis of a data set.
Draw the statistical test table.
What is a null hypothesis?
- A null hypothesis states there is no difference between the conditions.
- It states results are due to chance and are not significant in terms of supporting the idea being investigated.
What is probability?
Probability estimates the % chance of an event occurring.
What does significance level mean?
The level of probability at which it is agreed that the null hypothesis will be rejected, and the alternative (experimental) hypothesis accepted.
Wundt’s new ‘scientific’ approach to psychology was based on two major assumptions. What are they?
- All behavior is seen as being caused (determined)
- If behavior is determined, then it should be possible to predict how human beings would behave in different conditions (predictability)
What is falsifiability?
● This concept takes us firmly into the realms of philosophy of science.
● In the 1930s karl popper asserted that the key criterion of a scientific theory is falsifiability.
● Genuine scientific theories, popper suggested, should hold themselves up for hypothesis testing and the possibility of being proven false.
● He believed that even when a scientific principle had been successfully and repeatedly tested, it was not necessarily true.
What is a paradigm and who proposed this?
Thomas Kuhn suggested that what distinguishes scientific disciplines from non-scientific disciplines is a shared set of assumptions and methods.
Name the six sections of a psychological report.
- Abstract
- Introduction
- Method
- Results
- Discussion
- Referencing
When designing a study question, what should you make sure to include?
- 12/16 marks
- Use of a stem
- Requires broad understanding of research methods
- Be bold and creative!
- No entirely wrong answers
- All about justification
- Use subheadings
- You can say “i”
