RESEARCH METHODS AS COMPLETE STACK Flashcards
What are the 4 primary codes of the BPS?
Respect
Competence
Responsibility
Integrity
How do ethics community decide whether a research project should happen?
Ethics communities weigh up cost-benefit analysis and take into account the reputation of psychology.
What are Ethical Issues?
Ethical Issues arise when there is a conflict between the rights of the participant and the steps taken to reach the goals of the research.
What are Ethical Considerations to take into account?
- Consent
- Deception
- Confidentiality
- Debriefing
- Right to withdraw
- Protection from all harm
What is the general formula for a consent form?
-Purpose of study
-What happens
-Time needed
-Right to withdraw
-Any harm or help given
TICK BOXES
-Have read and asked any questions and understand rights to withdraw.
-Consent
What is debriefing?
Debriefing = AFTER the experiment, participants are informed of:
-Their data and rights around it
-If any deception occurred
-The results collected
What is the difference between a general population, target population and sample?
General Population
= Big, generalised group of people within an area.
Target Population
=Group of people who are the focus of the research project.
Sample
= Small group of people from the target population that are chosen to take part in the project.
What is the difference between bias and generalisation?
Bias is when certain groups are over or under represented, whereas generalisation is when findings from a study can be applied to a larger population.
What is Random Sampling?
Random Sampling is randomly selecting participants.
What are the advantages of Random Sampling?
ADV - free from researcher bias, all participants have equal chance of being chosen and should be fairly representative.
What are the disadvantages of Random Sampling?
DISADV - difficult and time consuming as participants might refuse to take part. May still be unrepresentative.
What is Systematic Sampling?
Selecting every nth member of the target population to take part
What are the advantages of Systematic Sampling?
ADV - Free from researcher bias and fairly representative
What are the disadvantages of Systematic Sampling?
DISADV - Every person doesn’t have an equal chance at being selected, time consuming and participants might refuse to take part.
What are the advantages of Opportunity Sampling?
ADV - Cost efficient and easy to conduct.
What is Opportunity Sampling?
Selecting anyone who is willing or available at the time.
What are the disadvantages of Opportunity Sampling?
DISADV - Open to researcher bias and fairly unrepresentative.
What is Volunteer Sampling?
Participants self selecting (volunteering) to take part.
What are the advantages of Volunteer Sampling?
ADV - Easy and minimal input from the researcher. Participants more engaged as they chose to take part.
What are the disadvantages of Volunteer Sampling?
DISADV - Open to volunteer bias and demand characteristics. Volunteers tend to be helpful, outgoing people so will be fairly unrepresentative.
What is stratified sampling?
Definition: Where the composition of the sample reflects the population of people in in certain subgroups - this represents the wider population.
What are the advantages of Stratified Sampling?
ADV - Avoids researcher bias and produces fairly representative sample.
What are the disadvantages of Stratified Sampling?
DISADV - Not a perfect sample and cannot reflect the way in which all groups are different. Time consuming.
What is the formula for stratified sampling?
(Sample size ÷ Population size) x Number of participants in sub group.
EXAM QUES
A researcher wants to study a sample size of 70 students from different subjects in college:
Latin - 145 total students
Spanish - 121 total students
German - 198 total students
French - 186 total students
Using stratified sampling, decide how many students should be used from each subject for a fair representation.
FORMULA
(Sample size ÷ Population size) x Number of participants in sub group.
Population size = 145 + 121 + 198 + 186 = 650
Sample size = 70 (states it in question)
(70/650) x 145 = 16 Latin students
(70/650) x 121 = 14 Spanish students
(70/650) x 198 = 21 German students
(70/650) x 186 = 21 French students
What are variable and what are the two main types of variables?
Variables = The things that you are investigating that changes
TYPES:
-Independant Variable = Variable that is manipulated by researcher or changes ‘cause’
-Dependant Variable = Variable that experimenter measures ‘effect.’
What does it mean when there are levels to the Independant Variable?
When the Independant Variable has different levels, this means there are different experimental conditions - a hypothesis will be written about these IVs.
For example - Group 1 drinks water whilst Group 2 drinks coffee - the ‘levels’ to the IV are the drinks.
What are extraneous variables?
‘Nuisance variables.’ These cause an effect on the DV which means the results are not accurate. These need to be controlled.
What are confounding variables?
These are a type of Extraneous Variable. They are found AFTER the data has been collected and HAVE had an EFFECT on the results.
What are participant variables?
Independant Variables related to participants that can effect results - such as eye colour, personality, etc (things in the participants that are DIFFERENT.)
What are situational variables?
Independant Variables related to participants that can have an effect on results - such as the environment, temperature, day, time of day, etc.
What are researcher variables?
Independant Variables related to the reseatcher that can affect resulst - such as body language, tone of voice, type of questions, etc.
What are controls?
Things you do to stop extraneous variables from affecting the overall results of the study.
What is operationalisation?
Taking something that is qualitative (not measureable) and making it clearly defined and measureable.
What are Demand Characteristics?
Demand Characteristics = Cues from the researcher which may lead to changed behaviours from the participants - angel / devil. Leads to inaccurate results.
What is the Social Desirability Bias?
Participants changing behaviours to be seen as more normal and socially desirable.
What are Investigator Effects?
Things that affect the study due to researcher behaviours - e.g. body language. This might be because they want to prove the hypothesis right.
How can you overcome Investigator Effects (EXPLAIN.)
Overcome by RANDOMISATION - this is use of chance methods during the processes of the experiment to reduce any biases. For example, using random sampling to select participants.
One example of randomisation is random allocation - what is random allocation?
Randomly allocating participants to the groups / conditions. This prevents bias and all participants have equal chance to be in all groups.
What are order effects?
Order effects are when the order of tasks participants take part in affects the study results. For example, taking part in a physical task first would mean participants might be too tired to concentrate on the next task and this effects results,
How can you overcome Order Effects? (EXPLAIN)
Overcome with COUNTER- BALANCING. This would mean that different groups take part in each conditions at different times. (REMEMBER ‘ABBA’ Group 1 A then B and Group 2 B then A)
What is standardisation?
Standardisation is the process by which all procedures and actions in an experiment is kept the SAME.
Give 5 examples of how you could standardise an experiment?
- Same time of day
- Same room / place of experiment.
- Same questions being asked
- Same temperature of surroundings
- Given identical standardised instructions.
What are researcher aims?
General statement of what the researcher intends to investigate.
What are researcher hypotheses?
Clear, precise, testable statements which state the relationship between variables to be investigated.
What is the difference between alternate and null hypotheses?
BOTH are a statement of prediction about your research, however alternate hypotheses say the research WILL yield significant results whilst null hypotheses say your research WILL NOT yield significant results.
Before you start a research, which types of hypotheses will you write?
BOTH alternate and null hypotheses are written, however at the end, only ONE is accepted.
What are Experimental Hypotheses?
A type of alternate hypothesis. Used for causal relationships. Predicts a significant difference or effect.
What are Correlational Hypotheses?
A type of alternate hypothesis. Used for non-causal relationships. Predicts a significant relationship or correlation.
What is the difference between Experimental and Correlation Hypotheses?
Experimental = causal, Correlational = non causal. In the question words such as ‘difference’ and ‘effect’ means Experimental Hypothesis whilst words like ‘link’ ‘correlation’ ‘relationships’ is correlational.
What is a Directional Hypothesis?
‘One tailed’ States the direction of the difference, relationship, etc.
Give an example of a Directional Hypotheses?
Women will score higher than men on a memory test.
What is a Non-Directional Hypothesis?
‘Two tailed’ Does not state the direction of the difference, relationship, etc.
Give an example of a Non-Directional Hypotheses?
There will be a significant difference in memory scores between women and men.
What is the difference between experimental method and design?
METHOD = Type of experiment being used e.g. Lab, field, quasi.
DESIGN = How the participants are used by researchers.
What is a Laboratory Experimental design?
A strictly controlled experimental method in which the IV is manipulated by researcher and DV is measured strictly.
What are the advantages of Lab experiments?
ADVANTAGES - High control over extraneous variables, replicable and standardised. Clear effect of IV on DV.
What are the disadvantages of Lab experiments?
DISADV - Lacks ecological validity and might lead to demand characteristics which means its less accurate.
What is a Field Experimental design?
Mimicks natural experimental design but isn’t fully natural as IV is still manipulated by researcher and DV is measured.
What are the advantages of Field experiments?
ADV - Higher ecological validity due to natural setting. Less demand characteristics.
What are the disadvantages of Field experiments?
DISADV - Difficult to control extraneous variables and more ethical issues are present.
What is a Natural Experimental design?
Full natural setting in which IV naturally happens and would have always happened (no manipulation from researcher.) DV measured by researcher.
What are the advantages of Natural experiments?
High external validity means more accurate results. Can research rare opportunities / IVs we don’t manipulate.
What are the disadvantages of Natural experiments?
DISADV - Needs a naturally occurring IV that are rare and hard to replicate. Lacks control over other variables.
What is a Quasi Experimental design?
Almost an experiment. IV is naturally occurring and recorded by researcher. Groups are decided in a non random manner.
What are the advantages of Quasi experiments?
Quite controlled which means more accurate. Any changes to DV is due to IV. Useful when ethics get in the way of true experiments.
What are the disadvantages of Natural experiments?
May have confounding variables which will affect the results.
What is an Independant Groups experimental design?
- Each participant only takes part in one condition and these results are compared with other participants.
What are advantages of Independant Groups experimental design?
ADV - Order effects less likely, less likely to guess aims so less demand characteristics.
What are disadvantages of Independant Groups experimental design?
DISADV - Needs more participants which means more costly / time consuming. Participant variables likely.
What is a Repeated Measures experimental design?
All participants take part in all conditions and their own personal scores are compared
What are advantages of Repeated Measures experimental design?
ADV - Participant variables are controlled - increases validity. Fewer participants so more cost efficient and cheaper.
What are disadvantages of Repeated Measures experimental design?
DISADV - Repetition and lots of tasks might lead to order effects / Demand Characteristics.
What is an Matched Pairs experimental design?
Each participant is matched with another participant, who shares relevant variables e.g age.
What are advantages of Matched Pairs experimental design?
ADV - No order effects, participant variables controlled, less likely to guess aims and so less demand characteristics.
What are disadvantages of Repeated Measures experimental design?
DISADV - Time consuming and much less efficient. Can’t fully contol participant variables.
What are the problems of Independant Groups Design and Repeated Measures Designs and how can you overcome these?
Independant Groups Designs have participant variables - overcome by random allocation Repeated Measures Designs have order effects - overcome by counterbalancing.
Give two examples of a self-report techniques?
Questionnaires.
Interviews.
EVALUATE Self-Report Methods.
+Detailed, free to express
-Subject to social desirability bias and lacks validity.
What type of questions make up questionnaires? (2)
Open and closed questions.
OPEN - allow as much detail as needed.
CLOSED - pick from pre-determined options.
EVALUATE open and closed questions.
OPEN
+Qualitative and rich in detail
+Participants feel less restricted
-Difficult to analyse and compare; often misinterpreted
CLOSED
+Easier to analyse and compare
- Limits details and participants feel frustrated.
What are the 6 types of closed questions?
- Forced / Fixed choice
- Likert Scale
- Rating Scale
- Ranking Scale
- Checklists
- Semantic differential rating scale.
What are forced or fixed questions?
A type of closed question.
Have to choose from a list of options.
Give strengths of forced or fixed questions.
+Easier to analyse and compare
and this is time efficient.
Give weaknesses of forced or fixed scale questions.
-Data lacks detail
-Participants feel forced to pick options.
What are likert scale questions?
A type of closed question.
Participants indicate how far they agree or disagree with something.
Give strengths of Likert scale questions.
+Allows for slight more detail whilst data remains easy to analyse and compare
Give weaknesses of Likert scale questions.
-Still lacks detail
-Participants might not agree too much to seem less extreme (central tendency.)
What are rating scale questions?
A type of closed question.
Participants give ratings on their own personal opinions.
Give strengths of rating scale questions.
+More insight / detail
+Still quite comparable
Give weaknesses of rating scale questions.
-Participants might interpret the ranks differently
-Might stick to middle to avoid looking extreme (central tendancy)
What are ranking scale questions?
A type of closed question.
Participants rank options.
Give strengths of ranking scale questions.
+Provides much more detail
+Comparisons can still be made
Give weaknesses of ranking scale questions.
-No options to rank 2 as equal, lacks validity
-Responses might be biased towards the values they see first -Lacks detail
What are checklist questions?
A type of closed question.
Allow participants to answer by choosing multiple options from the pre-determined choices.
Give strengths of checklist questions.
+Easy to analyse and compare
+Can select multiple options so more valid.
Give weaknesses of checklist questions.
-Data has a narrow range and lacks detail
-Difficult to interpret
-Participants might feel forced to select an answer.
What are semantic differential rating scale questions?
A type of closed question.
Questions which allow participants to indicate attitudes towards something on a scale between opposite words.
Give strengths of semantic differential rating scale questions?
+Easy to analyse and compare
+Allows for much more detail
Give weaknesses of semantic differential rating scale questions?
-Participants might interpret scales differently
-They may stick to middle to prevent seeming extreme
-Central tendancy bias
If you are conducting a questionnaire, what should you do to ensure that ethics are kept in line?
-Brief participants at start.
-Make the purpose clear.
-Make their data protection rights clear / right to withdraw
-Make sure questions are not ambiguous or leading
-Use simple language.
If you are conducting a questionnaire, what should you NOT do?
-Do not have overlapping choices or too few choices.
-No personal details unless absolutely necessary.
-No technical, confusing terms.
Give strengths of questionnaires.
-Quick
-Cost efficient
-Less demand characteristics, social desirabillity bias compared to interviews.
Give weaknesses of questionnaires.
-Low number of responses
-Questions could be misinterpreted = less valid
-Social desirabillity bias.
EVALUATE interviews in general as a whole.
+Participants free to express
+More detail
-Subject to social desirability bias -Time consuming
-Interviewers need to be trained which is costly
-Investigator variables
Describe structured interviews.
-Pre-determined questions
-Pre-determined order of questions
Give strengths of structured interviews.
+Relevant, comparative data
+Reliable and replicable
+Objective analysis
Give weaknesses of structured interviews
-Participants unable to expand answers = frustration.
-Lacks ecological validity.
Describe semi-structured interviews?
-Pre-determined questions and order BUT researchers can ask more questions to develop the data further.
Give strengths of semi-structured interviews.
+More detail and relevant data
+Free expression
Give weaknesses of semi-structured interviews
-Requires highly trained interviewers = COSTLY
-Only replicable to an extent
-Hard to interpret and compare
Describe unstructured interviews?
-Conversational style
-No set questions
-General aim is explored
-Interviewer encourages answers to be developed.
Give strengths of unstructured interviews.
+Less likely to show demand characteristics.
+Ecologically valid data
Give weaknesses of unstructured interviews.
-Need highly skilled interviewer so costly
-Not possible to replicate so unreliable
-Data difficult to compare
-Time consuming.
What are Naturalistic Observations?
Naturalist = Watching and recording behaviour in a setting where it would always occur.
What are strengths of Naturalistic Observations?
+Realistic environment means externally valid and realistic data
What are weaknesses of Naturalistic Observations?
-Less control over variables so replication is difficult.
What are Controlled Observations?
Watching and recording behaviours within a structured environment.
What are the strengths of Controlled Observations?
+Can focus on particular behaviours = replicable
+Controlled and standardised so only IV should affect DV.
What are weaknesses of Controlled Observations?
-Unnatural, so may lead to demand characteristics
-Social desirability bias
-Less external validity
Give a clear difference between Naturalistic and Controlled via an example.
Naturalistic - observing and analysing animals in the wild.
Controlled - observing and analysing animals in the zoo
What is an Overt Observations?
Participants behaviours observed and recorded with their consent or knowledge.
What are the strengths of Overt Observations?
+Reduced ethical issues as you have consent
What are the weaknesses of Overt Observations?
-Participants more likely to change their behaviours.
-Lacks validity
What is a Covert Observation?
Participants behaviours observed and recorded without their consent or knowledge.
What are the strengths of Covert Observations?
+Participants behaviour reflects their natural behaviour more.
+Increased external validity
What are the weaknesses of Covert Observations?
-More ethical issues are present e.g. invasion of privacy.
How can you overcome ethical issues in Covert Observations?
Use public spaces where CCTVs would already capture behaviours
Ask for retrospective consent after the study.
What are participant observations?
When the researcher becomes part of the group they are studying.
What are strengths of participant observations?
+Participants are unaware of being observed so behaviour is more natural - externally valid.
+Researcher can develop answers and ask questions for clarity.
What are the weaknesses of participant observations?
-Higher chance of observer/ researcher bias and variables.
-Ethical issues e.g. consent.
-Difficult to observe / take part in the study at the same time.
-Researcher may lose focus.
What are non-participant observations?
When the researcher remains outside the group but observes and records behaviours.
What are strengths of non-participant observations?
+Much more objective and researcher likely to not lose focus.
What are weaknesses of non-participant observations?
-If participants are aware of being observed, behaviour may change (Demand Characteristics) -Researcher might misinterpret behaviours.
-Reduced ecological validity.
What are Structured Observations?
Identifies target behaviours (e.g. stretching, itching, etc) and observes these and systematically records them.
What are the strengths of Structured Observations?
+Collecting data is easier as you have a target behaviour.
+Data is easier to analyse.
What are the weaknesses of Structured Observations?
-Some behavioural categories may be irrelevant.
-May have missed some behavioural categories.
-Data might lack details.
What are Unstructured Observations?
The researcher will observe and record ALL behaviours in front of them.
What are strengths of Unstructured Observations?
+Provides more in depth, detailed results.
+Less likely to have irrelevant data.
What are weaknesses of Unstructured Observations?
-Produces qualitative data which is harder to analyse.
-Risk of observer and researcher biases.
-Researcher cannot observe and record all data at once - DIFFICULT.
Sometimes a clearly defined Predetermined System is used for a structured observation.
What is a predetermined system?
A predetermined system:
Splits the structural observation into clearly defined behavioural categories. E.G. Anxiety (structural observation) = shaking, stuttering, stutter, etc.
What are Sampling Methods?
The method of sampling DATA for a structured observation.
What are the two types of Sampling Methods/
- Time Sampling
- Event Sampling
What is Event Sampling?
Observing at all times.
Counting the number of times a behaviour / event happens.
Must not miss any behaviours.
What are the strengths of Event Sampling?
+Useful if the behaviour doesn’t happen often and would be missed with time sampling.
What are the weaknesses of Event Sampling?
-Quite complex as many behaviours are present.
-Easy to overlook and miss some behaviours.
What is Time Sampling?
Using pre-determined time intervals- e.g. every 5 mins.
Every 5 minutes record and observe what happens.
What are strengths of Time Sampling?
+Effective in reducing number of observations that must be made TIME EFFICENT.
What are the weaknesses of Time Sampling?
-Behaviours can often be missed between intervals.
-Less representitive data.
What is a correlation?
Used to investigate associations, links and relationships between two co-variables.
What are Correlational Hypotheses?
-Uses existing, intervening data. -No variables manipulated.
-Predicts a LINK.
-The hypothesis can be a NULL or RESEARCH hypothesis.
What are Null Correlational Hypotheses?
States that there are NO correlations or links between the variables.
What are Research Correlational Hypotheses?
States that there is some form of correlation or link between the variables.W
What does a positive correlation mean?
As one variable increases, so does the other.
What does a negative correlation mean?
As one variable decreases, the other increases.
What does a zero correlation mean?
No correlation between the two variables.
What is a correlation coefficient?
Shows how strong a correlation is between the variables.
Describe the levels of the correlation coefficient.
1 / -1 = Perfect correlation.
0.7 < R < 1 OR -0.7 < R < -1 = Strong correlation.
0.3 < R < 0.7 OR -0.3 < R < -0.7 = Moderate correlation.
0 < R < 0.3 OR 0 < R < -0.3 = Weak correlation.
0 = Zero correlation.
What is a Pilot Study?
-A small scale version of an investigation carried out beforehand.
-Similar to a trial run.
What are the aims of pilot studies?
-Checks if procedures, measurements, etc are correct.
-Identifies other mistakes and quick fixes that can be made.
When are pilot studies used?
-Used beforehand in investigations and self reports.
Give an example of a scenario where pilot studies are used?
E.G. A questionnaire is sent out to 50 people before sending to 500. Some questions are ambiguous, so the researcher changes the wording of the questions to make it clearer.
What is a single blind trial?
Participants are unaware of the aims and other aspects of the experiment. This is done to control demand characteristics.
What is a double blind trial?
Neither the researcher nor participant knows aims of study. REDUCES investigator effects and biases.
What is Peer Review?
The assessment of scientific work by other specialists in the same field to ensure the research is of high quality.
What are the 3 aims of Peer Review?
- Allocates research funding to those who deserve it most.
- Validates the quality and relevance of research.
- Suggests easy fixes and amendments before publishing.
What are strengths of Peer Review?
+It ensures all data is accurate and appropriate research receives appropriate funding.
+Maintains the reputation of psychology.
+It is ANONYMOUS so produces objective, honest reviews.
What are weaknesses of Peer Review?
-Anonymous reviews might mean more critical / harsh.
-Researchers and reviewers are usually in competition for funding so may be more critical. -Some journals only publish headline grabbing research IGNORING other research.
-Researchers may ignore research that goes against long, old-term research which slows down the rate of change.
What is the economy?
The state of a country or region in terms of the production and consumption of goods and services.
What is meant by the ‘Implications of Psychology on the Economy?’
Psychologists must take into account the effects that a research will have on the economy - a good or bad implication.
Give a scenario where psychology had a good implication on the economy?
Mental Disorder Research.
Improving treatments of mental disorders would lead to more people working and less people being unemployed or on paid leave. BOOSTS economy.
Give a scenario where psychology had a bad implication on the economy?
Attatchment Research.
This explained that fathers ALSO need to build an emotional connection with their children. This led to paternity leave, which meant more people were on paid leave.
What is Quantitative Data?
Numerical data - expressed in numbers.
What are the strengths of Quantitative Data?
+Easy to analyse and compare.
+More objective
+Less open to bias.
What are the weaknesses of Quantitative Data?
-Lacks details
-May be misleading
-May be invalid and not represent real life.
What is Qualitative Data?
In depth, descriptive data such as words and images.
What are the strengths of Qualitative Data?
+Provides more detail, opinions and thoughts.
+Increased external validity.
What are the weaknesses of Qualitative Data?
-Harder to analyse / subjective.
-Expensive and time consuming. -Harder to compare
What is Primary Data?
Data collected for a study is directly collected by the researcher.
What are strengths of Primary Data?
+Authentic data
+Fits aims of study
What are the weaknesses of Primary Data?
-Time consuming
-Expensive to obtain
What is Secondary Data?
Data which is collected by other sources and already exists. It should link to the study in some way.
What are the strengths of Secondary Data?
+Less time consuming
+Inexpensive
+Easily accessed.
What are weaknesses of Secondary Data?
-May be outdated
-May not match or fully link to your research projects.
What is Meta Analysis?
The method of looking at many studies and selecting a hypotheses and aims which link to all the studies. Using this, after analysing, a joint conclusion is produced.
What is meant by Descriptive Statistics?
Use of tables, graphs, summary tables to represent and analyse data while showing trends.
What is meant by Measures of Central Tendancy?
A general term for any measure of average value in a data set.
What is meant by Measures of Dispersion?
A general term for any measure which shows spread and variation within scores.
What are 3 main examples of Measures of Central Tendancy?
- Mean
- Median
- Mode
What are 2 main examples of Measures of Dispersion?
- Range
- Standard Deviation.
What is meant by Mean?
An average of the data.
What are the strengths of calculating a mean value?
+Each value is accounted for and represented.
What are the weaknesses of calculating a mean value?
-It is easily affected by outliers.
What is meant by Median?
The middle score of all the ranked data.
What are the strengths of calculating a median value?
+Less affected by outliers and skewed data.
What are the weaknesses of calculating a median value?
-Middle value doesn’t represent all the values / data as a whole.
What is meant by Mode?
The most common score among all the data.
What are the strengths of calculating a modal value?
+Useful with categorical and qualitative data to analyse.
What are the weaknesses of calculating a modal value?
-Modes are not unique. Sometimes more than one modal value is present.
-The mode relies on a value repeating and this isn’t always the case.
What is meant by Range?
-One example of a measure of dispersion which shows the spread of data.
What are the strengths of calculating a range?
+Quick and easy to calculate
+Time efficient.
What are the weaknesses of calculating a range?
-Vulnerable to outliers and extreme scores which means misrepresented data
What is meant by Standard Deviation?
A sophisticated measure of dispersion that tells us how data is spread over the mean value.
What are the strengths of calculating a standard deviation?
+Is much more accurate at showing how data is distributed +Also shows distribution around mean so more detailed.
+Less affected by outliers.
What are the weaknesses of calculating a standard deviation?
-Doesn’t show full range.
-Hard and time consuming to calculate.
-Can only be used when an independant variable is measured against a frequency.
-Assumes a normal distribution which isn’t always the case.
What is meant by a high standard deviation?
High Standard Deviation = Data is more spread out over the mean.
What is meant by a low standard deviation?
Low Standard Deviation = Data is much more clustered tightly around the mean.
What are the 4 main ways to present and display quantitative data in psychology?
- Summary Tables
- Bar Charts
- Scatter Graphs
- Histograms.
What is meant by a normal distribution?
-Heavy on the middle, low on the sides, showing few outliers.
-Mean, median and mode occupy same midpoint.
What is meant by a positively skewed distribution?
-Heavy on the left.
-Mode at highest point, followed by median then mean.
-Mean towards tail, as it is easily manipulated by outliers.
What is meant by a negatively skewed distribution?
-Heavy on the right.
-REST SAME AS NEGATIVELY SKEWED.
-Mode at highest point, followed by median then mean.
-Mean towards tail, as it is easily manipulated by outliers.
What is probability?
The likelihood of an event happening.
What is significance?
A statistical term that indicates the strength of correlations and helps decide which hypotheses to accept or reject.
What is meant by the ‘Standard level of significance?’
This is an indication as to which research is acceptable. Psychologists usually accept research at a probability of 0.05 or below - this means there is LESS than 5 in 100 probability that results occurred due to chance.
In what instance might the 5% rule not apply?
When you need results to be definite. For example, during drug trials you may have to have a stricter 1% rule.
What would you do if the standard significance level / probability is ABOVE 0.05?
Probability of results occurring due to chance is higher that 5%. The Alternate hypotheses is rejected and the Null is accepted.
What would you do if the standard significance level / probability is BELOW OR EQUAL TO 0.05?
Probability of results occurring due to chance is less than or equal that 5%. The Alternate hypotheses is accepted and the Null is rejected.
What is meant by a Type 1 Error?
An error that occurs when a researcher says a study is significant (P<0.05) and accepts the alternate hypotheses BUT in reality the research is not significant.
Should have accepted NULL and rejected ALTERNATE.
Give a scenario where a Type 1 Error has occurred?
A pregnancy test showing a positive result when the woman isn’t actually pregnant. ‘False positive result.’
What is meant by a Type 2 Error?
An error that occurs when a researcher says a study is not significant (P>0.05) and accepts the null hypotheses BUT in reality the research is significant.
Should have accepted ALTERNATE and rejected NULL.
Give a scenario where a Type 2 Error has occurred?
A pregnancy test showing a negative result when the woman actually is pregnant. ‘False negative result.’
Why might Type 2 Errors occur?
Too strict accepted level e.g. only accepting if the results are 0.01.
What is meant by levels of data?
Types of data being ordered and ranked depending on a range of factors - each level is more sophisticated than the previous.
What are the 4 levels of data, starting at the lowest level?
- Nominal
- Ordinal
- Interval
- Ratio
What is nominal data?
Lowest level of data.
Categories split into subgroups and arranged as a frequency.
What are the strengths of nominal data?
+Some data can ONLY be recorded as nominal
What are the weaknesses of nominal data?
-Nominal numbers are just labels - no numerical value.
-Vague as you don’t know nothing about data other than the numbers.
What is ordinal data?
Second level of data.
Data placed in a specific order which is then ranked.
What are the strengths of ordinal data?
+Allows to judge the magnitudes such as worst and best.
What are the weaknesses of ordinal data?
-Can only see the ranking and not the differences in proximity between rankings.
What is interval data?
Third level of data.
Data expressed by a standard measure / unit which has equal intervals.
Can contain minus numbers.
What are the strengths of interval / ratio data?
+Allows to judge magnitudes such as who is best and worst.
+Can make meaningful comparisons between individual pieces of data.
What are the weaknesses of interval / ratio data?
-Highest level of data and has NO WEAKNESSES.
What is ratio data?
Highest level of data.
Data expressed as a score, but has an absolute 0 (cannot contain minus numbers).
Doesn’t have standard equal intervals.
Which measure of central tendancy do you use for each level of data?
- Nominal = Mode
- Ordinal = Median
- Interval / Ratio = Mean
What is statistical testing?
A way of determining and testing which hypotheses should be accepted or rejected.
What is the sign test?
Type of statistical test which tests for differences in specific data.
Type of non-parametric test.
What are the 3 things needed for the sign test?
- Nominal data
- Research studying differences 3. Must be related - repeated measures or matched pairs.
What are the strengths of the sign test?
+Easy method
+Doesn’t require normal distributions
+Can test BOTH hypotheses.
What are the weaknesses of the sign test?
-If an observation is equal, or shows no change, the data is ignored.
-Important data overlooked.
-Non-parametric test means reduced statistical power compared to other tests.
What are parametric tests?
Powerful test which is highly likely to detect differences and correlations.
What is the criteria for parametric tests? (5)
Data must not be skewed.
Data must have normal distribution.
Data must be interval / ratio.
Data must have no outliers.
Data must have similar standard deviations.
Which test would you use for unrelated nominal test?
Chi-Squared Test.
Which test would you use for unrelated ordinal test?
Mann Whitney U Test.
Which test would you use for unrelated interval test?
Unrelated T-test.
Which test would you use for related nominal test?
Sign Test
Which test would you use for related ordinal test?
Wilcoxen Sign Test
Which test would you use for related interval test?
Related T-test.
Which test would you use for a correlating nominal test?
Chi-squared test.
Which test would you use for a correlating ordinal test?
Spearmann’s Rho Test
Which test would you use for a correlating interval test?
Pearson’s Correlation Coefficient.
What is meant by related test?
Matched Pairs or Repeated Measures designs.
What is meant by unrelated test?
Independant Groups Designs.
Which tests must be greater than or equal to the critical value in order to be significant? (5)
-Chi-Squared Test
-Unrelated T-Test
-Related T-Test
-Spearman’s Rho
-Pearson’s Correlation
All the tests containing the letter ‘R’
Which tests must be less than or equal to the critical value in order to be significant? (3)
-Mann Whitney U-Test.
-Sign Test
-Wilcoxen Sign Test.
