Flashcards in Research Methords Deck (83):
What are the four types of experimental methods
What is a Laboratory experiment
Highly controlled conditions. The independent variable is manipulated by the researcher and the participants are assigned randomly to each condition
Explain 2 strengths of a lab experiment and 2 limitation
The high level of control over valuables means of research can easily be replicated this enables others to see if the findings are reliable.
The high level of control also means are still there is likely to be internally valid, it is unlikely that a trainee is very bells will confound these results. Therefore we can establish cause and effect relationship between the independent variable and the dependant variable.
What is a field experiment
- natural environment of those being studied
- researcher still manipulates the IV
Name a strength and 2weakness of a field experiment
Improves ecological validly therefore it is possible to generalise findings because participants are in there natural setting
Participants are unaware they are taking part they are not able to give informed consent- against ethical implications
- lower level of control - difficult to establish control over extraneous variables, difficult to replicate and we are less likely to establish a clear cause and effect relationship
What is a natural and a quasi experiment
Natural - the group is pre existing eg- brain damage. So the researcher can have no control over who is in each condition. However the DV can be measured under laboratory conditions.
Quasi- when he IV is a difference that exists between people eg gender. Random allocation isn't possible as the IV is a characteristic of the pope on. DV can be measured under lab conditions
Give 2 strengths and 2 weakness of a natural/quasi experiment
- allows us to study areas which would be unethical for an experimenter to manipulate the independent variable
- more ethical as effects of the IV have occurred naturally and the researcher has not altered ps behaviour
- the situation may rarely occur so there maybe limited research opportunities to replicate the study and establish reliability
- cannot demonstrate causal relationship because the IV has not been directly manipulated and there could be confounding variables that can't be controlled since random allocation of ps isn't possible.
What is the difference between a directional and non direction hypothesis
What do hypothesis for correlation always predict ?
Direction (one tailed )
Predicts the direction in which the results are likely to occur eg - more or les
Non direction (two tailed )
Does not predict and expected direction - rather a difference between the two conditions.
- always predict a direction and a strength of relationship eg- there will be a strong positive relationship between hour spent practicing and score test.
Along side a directional or non direction hypothesis wha other hypothesis will also be created
A hypothesis - this predicts no difference or relationship between variables but suggests that any difference between two conditions is due to chance
What is a variable
What is the difference between the IV and the DV
Anything that can change it can be physical e.g. Temperate or be a quality eg - aggression
It is a measurable characteristic that can differ from one observation to the next.
IV- manipulated by the reacher
DV- the variable measured by the researcher
Why is a control group used
As a baseline comparison group - the experimenter will see if the effects happen in the experimental condition but not in the control. If this is the. Are the researcher can be sure that what they thought caused the effect does indeed do so.
What is operationalised
What does it allow
Variables are defined so that they can be measured.
- allows other researcher to replicate research to see if results are reliable.
What are extraneous variables
What are they called when they effect the results ?
Name the three types of extraneous variables
Something unpredictable that happens in the researcher than could of effected the IV which wasn't the DV.
Called - confounding variables
Participant variable - age, gender, IQ
situational variables - noise level, room layout
Experimental variables- experimenters appearance , behaviour, gender
Where can unwanted variables come from
Random error or constant error
What are random errors
Now are they attempted to be controlled for
Ps state of mind
However there unlikely to have a huge impact as it is unlikely that all ps in one condition will be tired.
- controlled through random allocation
How are constant errors controlled
What is a pilot study
A small scale trail run of the actual study with a smaller number of ps. Aim to find out if there are any problems with the design and measure the equipment being used.
It may reveal further clarification regarding the operationalisation of variables or other extraneous variables which exists but hadn't previously been considered.
Ps may also suggest improvements e.g.- if they had worked out the hypothesis and changed their behaviour or got bored.
What is an independent measures/ group design
Gave one advantage and one disadvantage ?
How is individual difference controlled for ?
Different ps are used in different conditions e.g.- one p is only in one condition.
Advantage - one p is only used once this minimises the chances of demand characteristics. So they are less likely to alter their behaviour so the experiment is more alternately valid.
Disadvantage - participant variables may confound results. Meaning we can not be as confident that it is the IV effecting the DV
- random allocation - can be done by writing names of ps on a piece of paper and pulling names from a hat - every other name picked out maybe allocated the condition one.
What is a repeated measures design
Give one advantage and 2 disadvantages
What is used to control order effects
Exposing every participant to each experimental condition therefore ps are being tested against themselves
Advantage - participant variables won't effect the results as each p acts as there own control. Can be more confident in establishing the cause and effect relationship
Disadvantage - order effects - participation in one condition affects performance in another. Ps may improve through practice this then reduced the validly of the study
- demand characteristic as their exposed to both conditions maybe able to work out what behaviour is expected and change their behaviour accordingly.
Counter balancing- equal number of ps undertaking the required tasks in different orders e.g.- half do task a first and half do task b first and swap.
What is a matched pairs design
Give 2 advantages and 2 disadvantages
Matching two ps and putting one in condition one and the other in condition two.
May match gender or IQ .
Ad- same stimulus can be used for both conditions removing this as a confounding variable
- demand characteristics are minimised
Dis- participant variables at it is unlikely that 2 ps match on all characteristics. Even if they are the same age and gender they may have different levels of motivation.
- large number of ps needed may not be possible to use everyone who agrees to take part if a match can't be found
What is a correlation
A collection of statistical techniques which can assess the direction and strength of a relationship between two measures.
Establishes a relationship.
Doesn't manipulate either variables.
What can be calculated in correlation research
The correlation coefficient
This is a numerical measure of the strength of the rel between variables
Varies between -1 (perfect negative correlation) to +1 (perfect positive correlation)
0 shows no correlation
Define a positive negative and zero correction
Positive- as the value of one. Arguable increased the value of the second variable increased. When one variable decreases so would the other.
Negative - one increase one would decrease
Zero- no relationship between the two variables
Why is the correlation method used
To assess the reliability of a test or measure
Test-retest reliability- a way of measuring the constituency of a psychological test. To determine the constituency of the test across time.
- early stages of research
Used as a pointer for further more detailed research. If the correlation is significant then it may be justified to design and carry out experiment study.
Name one strength and one weakness of a correlation
Strength - can be used when it's not ethical or practical to manipulate variables, so offers alternative research options.
Weakness- cannot establish a cause an effect relationship
What is the opposite of a controlled observation
Define both of them and give one advantage and one disadvantage
Controlled- the researcher attempts the control certain variables. Carried out in lab.
Ad- researcher can manipulate the situation to ensure the behaviours that they are interested in are happening
Dis - demand characteristics could mean a change in behaviour meaning the study would lack internal validity.
Naturalistic - examine behaviour as it occurs in a natural setting without investigator Interfering in anyway
Ad- behaviour are more likely to reflect natural behaviour meaning it's more ecologically valid
Dis- difficult to replicate as it is not standardised difficult to test for reliability
What is the opposite of participant observation
Define both of them and give one advantage and one disadvantage
Participant -researcher becomes activity involved in the situation being studied. They take part in the same activities and become a member of the group.
Ad- ps behaviour is more likely to be natural as the situation is lee formal therefore the validity of the results will be in rare. Researcher has first hand experience and therefore more of an insight into behaviour being studied.
Dis- difficult for the researcher to record info accurately if there taking part in the activities.
Non participant -the researcher remains external to those being observed n remains distant
Ad- remain objective as they are less likely to build a relationship with ps
Dis- researcher will not gain that much insight into the behaviours being studied reducing validity.
What is the opposite of a overt observation
Define both of them and give one advantage and one disadvantage
Overt- aware that the observer is studying them
Ad-more ethically appropriate, able o give informed consent.
Dis- demand characteristics - may try to behave socially desirable - e.g. the validity
Covert- do not know there being observed. Researcher must act as one of the group, if it is non participant they may watch from a distance
Ad- this may allow to study groups who are likely to refuse eg- deviant
Dis- unethical - deception, no informed consent.
What is the opposite of a structured observation
Define both of them and give one advantage and one disadvantage
Structured- already determined which behaviours are being observed and will use a standardised check list of behavioural categories to recall frequency
Ad- behavioural categories mean it can be easily replicated increasing the reliability.
Quantitive date which is easier to analysis
Dis- behaviour categories are unlikely to include all types of behaviour meaning some behaviours may not be recorded limiting the validity.
Unstructured - research records behaviour as it occurs. No plan on what to observe.
Ad- researcher isn't restricted in terms of which behaviour they record. Increasing the validity.
Dis- unreliable method because the observer may choose to record different info.
What should behavioural categories do/be
-cover all possible component behaviour
- by mutually exclusive
Give an example and definition of two sampling procedures
Time sampling- recording behaviour in a sry time frame eg- every 30 seconds
Event sampling - observing and recording a complex event every time it occurs eg- every time someone hits someone
What's intra observer reliability
Observers constantly code behaviour in the same way, so there is agreement between the observers, what one interprets as hitting so such the other.
What are self report techniques
Ps give in for about them-self. Questionnaires can be done face to face, by post, phone, internet or left for ps to collect from central points
What are the 2 types of questionnaires give ad and disads
- yes or no questions/ 1-10
Ad- quick and easy for ps - people are more willing to complete, increasing sample sizes.
Dis- ps may not have the option available that reflects the answer they want to give, decreases the validity as there forced to pic something that doesn't reflect the truth
P is free to express there own words. Qualitative date.
Ad- analysis of data is open to researcher bias as they have to interpret ps answers and classify them. This could be difficult of p uses slang or hasn't explained themselves properly
Give a strength and weakness of a questionnaires
Ad- they are standardised and can be easily replicated to test for reliability
Dis- the sample is likely to be biased against the people who have time to compete a questionnaire, have strong opinions on the topics and are able to read and write.
What are the four types of interviews
Structured, unstructured, clinical and semi structure
Define a structure interview
Give 2 ads and 2 disads
Quantitative data. The wording and order of the questions asked will be the same for every interviewee. Essentially a questionnaire over the phone.
Ad- less likely to deviate from the topic that is the desired focus of the interview.
Easy to replicate and test for reliability.
Dis- researcher can not follow up any new lines of enquiry that become apparent during the interview, meaning vital info may not be included
They may not revel their true thoughts and feelings.
What are unstructured interviews give ads and disads
Less rigid and very little.
Start with one or two set questions and then pick up on issues that the interview raises.
Ad- less for a, relationship of trust and understanding is established which may mean the interviewee is more willing to open up and discuss sensitive
Dis- fewer interviews can be taken out because they take longer- smaller samples - which will decrease the population validity of the results.
What is a semi structured no a clinical interviews
Semi structured interviews - most successful approach, supplemented by opportunities for the interviewee to expend there answers
Clinical- these are used in order to diagnose abnormal behaviour - carried out by psychiatrists. Follow up questions are given based on the answer
Giving the interviewer the opportunity to explore interesting or unexpected answers.
Give an advantage and disadvantage of interviews
Flexibly - can enable researcher to explore complex issues that may be difficult to investigate through questionnaires. The interviewer can tailor the response of an interviewee so that issues can be explored in depth
May give socially desirable answers.
What is a case study
In depth study of an individual or small group
Unique or extreme circumstances
Eg- effects of brain damage
Things that can't be studied experimentally
Multiple research methods may be used eg- interviews, observations and psychological tests in order to build up a detailed picture of the case. A number of sources can be used from individuals themselves.
Strength and weakness of case study (2)
Ad - large quantities of qualitative and quantitive data and therefore provide a detailed understanding of the behaviour in questions.
Case studies may provide new insights that test the applicability of a theory.
Not possible to make her realisation from the, or test reliability
Info is often only gathered after the data has happened- don't know what they were life before
What's internal validity ? How can it be accessed
What is external valid? How can it be assessed?
About consistency. For instance if you use an IQ test on someone twice, you should get the same answers twice. If a measure doesn't perform like this it's unreliable.
-refers to whether a measure is consistent within itself. Eg- if all the questions are worth 5 marks they should be of equal difficulty.
can be accessed using the split half method, this involves dividing the test in two and then comparing the results of two halves. If there is reliability there should be a very strong correlation between the two sets of Scores.
- refers to the consistency of measures over time if it is repeated. For example personality tests should not give different results if a person takes it on 2 occasions. To assess this a psychologists can use the test retest method. Do it on two separate occasions and compare the scores, again the correlation should be strong. You could also get another research to carry out the same test to check reliability, it checks for bias data collection. There will be a strong positive correlation if there is no bias and this is called getting a inter rater reliability.
How can validity be improved
By developing consistent forms of measurement e.g.- taking multiple measures and calculating the mean also standardisation.
Standardised instructions - carried out in a consistent manner.
Essentially it is improved through standardisation. Keeping the procedure the same for all ps.
What is validity
What is internal validity
What are some threats to internal validity
How can internal validity be improved
Refers to how accurately something measures what it claims to.
- internal - that the IV Is effecting the DV.
- threats - demand characteristics, researcher bias, participant variables, poor design, investigator effects.
- improved through controlling extraneous variables, double blind technique, use of counterbalancing if using repeated measures, random allocation in independent measures. Standardisation.
What is external validity
Whether you can generalise results beyond the specific research itself, other people, setting, time.
Ecological- whether the findings can be applied to other settings, this may be lacking if the study does not reflect a real life situation.
Population validity- refers to whether the finding from one sample can be applied to other groups.
Temporal validity - refers to whether the findings apply to different periods of time eg- are some of the studies that were conducted in 1970 still relevant today.
How can external validity be improved ?
How can it be accessed?
More naturalistic setting on a wide range of ps.
-face validity - this simply involves judging whether the test appears to measure what it is supposed to eg- if a questionnaire is measuring aggression the question should be around how angry you feel. If there asking where do you like to go on holiday it would lack face validity.
-Concurrent validity - correlating scores on a test with another test that is known for being valid. Eg- if you found a new was to measure how extravert someone was you could compare scores on Eysencks personality inventory.
What does a sample need to be
Representative of the target population- have the same proportionate characteristics. For if the population is 55% male then the same should be 55% male.
What is a volunteer sample
Ads and disads
People volunteer to take part. Researcher may put a notice up in a newspaper.
Ad- easy way of gathering ps, reducing cost and time of researcher.
Dis - likely to be biased of people who have the time or are more helpful.
What is a opportunity sample
Ads and disads
Asking whoever fits the characteristics of the target population. The psychologists may approach a near by institution such as a local school, gym or hospital, or they may stand in the town and ask by passers.
Ads- easiest methods because you just ask the first p you can find who are available and need the criteria which is quicker than other methods.
Disad- researcher bias - may approach people who look more friendly and helpful.
What is random samplying
Ads and disads
Each member of the target population have an equal chance of being selected. Eg - all names are put in a hat and those picked out are the ones used.
Ad - each person in the target population has an equal chance of being selected.
Disad- people drawn from the target population may not want to participate. Or those who are randomly selected my not represent the diversity of your target population. Unbaksed se,ction sies not garantee and unbiased sample.
What is systematic sampling
Ads and disads
Involves having a list of the target population and taking every nth person from the list. It is necessary to calculate the size of the population and then assessing what size of sample is needed to work out what the sampling interval is. If the population is 1,000 and a sample of 20 is needed then 1000 divided by 20 = 50. Therefore every 50th name from the list of the population will be selected to participate.
Ads- there is no bias so the sample should be representative
Disads - the method maybe impractical as it requires details of the target population which may be unknown.
What is stratified sampling
Ads and disads
This involves dividing a population into characteristics for the research e.g.- young females. Then a percentage of the population from each stratum is randomly sampled. This is a small scale reproduction of the population because of 28 % of the population is made up of young females then 28% of the sample will be selected from this group.
Ad- method is free from bias as the sample is selected randomly.
Dis- may be impractical as it requires detailed knowledge of the target population and its characteristics which may be unknown.
What do researchers who are carrying out research have to abide by
What are the four principles
BPS produced the code of ethics and conduct.
- respect, competence, responsibility, integrity.
What are the guidelines and how are they dealt with if broken?
Deal? Presumptive consent - a sample from the target populations are informed about the aims of the research and its design including the use of deception, if they agree that they would have consented knowing the true purpose then we can presume that others would feel the same.
Prior general consent - they know deception may occur but don't know in what way.
Retrospective consent - asking them after they have taking part. If they do not give consent there date will be destroyed.
Avoidance of deception - if deception occurs ps cannot give informed consent.
Deal? Debrief - reminded of the purpose of the investigation. Restore ps to the same state they were in before.
Protection from physical and psychological harm- harm should be no greater than everyday life.
Deal? Stop the study of you suspect harm is being caused. Provide support during debrief.
The right to withdraw - free to leave at any time and withdraw their data at the end.
Deal?be made aware of the right to withdraw before they start, if they wish to withdraw then the data should be destroyed.
Confidentiality- ps date should not be disclosed to anyone unless agreed in advanced. Results will be published anonymously.
Deal? Anonymity - details about someone's identity are not even collected in the first place. Could use a coding system whereby numbers correspond to names but this info should be kept
What is quantitive data
Ads and disads
Ads- more reliable, allows for comparisons to be made easier and statistic tests to be carried to test whether results are significant
Disad- lack detail so the research does not gain much insight into people's feelings.
What is qualitative data
Ads and disads
No numerical data- words, images, videos. Generated through open questions and case studies.
Ads- highly detailed providing an insight into people's thoughts and feelings
Disads- harder and more time consuming to analyse and may be open to interpretation by th researcher.
What is primary data
Ad and disads
Collected by the psychologists for the purpose of the study currently being undertaken
Ads- data collection can be designed so it fits the aim on the hypothesis
Disads - more time consuming and costly for the researcher as they have to design the study, seek out ethical approval and carry it out.
What is secondary data
Ads and disads
Secondary - data collected for a purpose other than the study. This may include research carried out previously by themselves or another psychologist or it may involve using government statistics.
Ads-quick and less time consuming
Disads- the data was not collected with the specific study in mind therefore it may not fit the purpose of the psychologists.
What is a meta analysis
Combing the findings of several studies into a particular research area.
Define nominal, ordinal and interval data
Nominal- lowest level. Count the number of times something happens. Data is in separate categories. Can't make quantitive comparisons between people.
Ordinal - put into order and ranked. Difference between people. Can't say someone is twice as sociable as someone else as the gaps between variables aren't equal.
Interval- allows us to make quantitative comparisons between people. Intervals are equal eg- temperate
What are descriptive statistics
Provide a summary of a set of date.
Include : measures of central tendency and measures of dispersion.
What are the measures of central tendency
Give one ad and one disad of all 3
Mode, median and mean.
Ad- quick and easy to calculate
Dis- less appropriate when there are multiple modes.
Ad- unaffected by extreme scores.
Dis- if there is an even number of data the median may be a value that is not even presented in the ordinal data.
Ad- includes all the variables in a set of data
Dis - because it takes into consideration all the values in the set of data, the mean is easily skewed by extreme values.
What are the measures of dispersion
Give ads and dis ads
Range and standard deviation
Ad- quick and easy to calculate
Dis- skewed by extreme measures
ad- allows us to interpret individual scores
Dis- more complicated to calculate.
When does a skewed distribution occur
When they are extreme scores.
Negative skewed distribution- when people receive high scores but there is an extreme low score.
Positive skewed distribution - people have low scores but there is an extreme high score.
When is the sign test used
How do you calculate the sign test
When there is
- a difference between two sets of scores is predicted
- the data is least normal
- repeated measures or matched pairs had been used.
1. If the experimental score is hight +
If it is lower -
If it is the same you don't use it.
2. Count the number of times the less frequent sign occurs this is then known as S.
3. Count the total number of pluses and minus this gives us N
4. One tailed or two tailed test
5. Look at the critical table. If the calculated value is equal to or less than the critical value then the result is significant.
What is a peer review
What do they act as ?
An independent scrutiny of scientific work by other who are experts in the field. It allows other researchers to scrutinise the quality of each other's research and contribute to the body of knowledge around the world.
Reviewers lass there comments back and may recommend:
- recommended for publishing
- recommend for publishing if minor modifications are made
- research may be rejected with the recommendation that the research makes major modification and resubmit the work
- the research may be rejected.
Act as a quality control mechanism, ensuring that only the highest quality scientific research enters the public domain. It may inform decisions about the allocation of research funding e.g.- medical research council will be more willing to fun research that is going to be carried out by a university or research team who have proved track records of carrying out h the quality research.
What are problems with a peer review
-they may not always be a suitable qualified expert to scrutinise research. This may result in poor quality research being published because the researcher didn't fully understand.
- single blind doesn't guarantee objectivity, peer reviews may be used as a mean of protecting their own status of their own theory, seeking revenge.
- publication bias - significant results are more likely to be published than non, this could hinder our understanding on human behaviour.
- studies that are challenged are unlikely to be published. This slows down the development of science as new ideas are not disseminated as quickly as they should be.
How can qualitative data be analysed
Content analysis and thematic analysis
What is content analysis
Give an ad and a disad
Form of observational research.
People are studied indirectly through the media they have produced.
Eg- verbal communication, written communication.
The aim is to summarise and describe the communication in a systematic way to draw conclusions.
Qualitative data is then quantified through the use of copying; these may include the number of times particular words, themes or characters appeared.
Ad- does not require direct contact with individual. Much of the material will already be in the pub,if domain and this means it is an ethical research method to use as the psychologists does not need to gain informed consent.
Dis- some researchers may argue it is inappropriate to convert qualitative date into quantitive date as it means that all the rich detail is lost and the final summary may not be a valid representation of the complexity of the original media.
What is thematic analysis
Ads and disads
Organising, describing and interpreting qualitative data. It involved identifying, Anaya's yin grandson and reporting themes that are recurrent in the media being studied.
Themes emerge from the data being studied rather than being imposed on the data due to pre existing theory's that the researcher may have.
Lengthy process that follows these stages:
Searching for themes
Defining and naming themes
Ad- better than content analysis as the data remains in its quantitative form and provided a much more detailed insight into behaviour being investigated.
Dis- time consuming process, the amount of data that can be analysed is limited, may involve the study of less then 10 individuals which limits the extent to which generalisations can be made.
How is psychology made more scientific
- objectivity and empirical methods
- replicability and falsifiability
-theory construction and hypothesis testing
- paradigms and paradigm shifts
- probability and significance
How does objectivity and empirical methods make psychology more scientific
- if the research is objective then they are not letting their own bias or expectation influence the research process. IMPORTANT as science aims to undercover the truths about the world therefore it shouldn't be open to interpretation. Also scientific research may have practical implication therefore much not be flawed and open to bias as this could have serious repercussions.
- behaviour should be observed and measured scientifically. Research should not be biased on thoughts , beliefs or reasoned argument. Free from bias.
How does replicability and falsifiability make psychology more scientific
-used the access the validly and the findings. If the outcomes are the same then this verifies the truth of the original results. Replicating research in different settings and with different populations allows us to draw conclusions about the generalisability of the results.
- in order for a hypothesis to be considered scientific it must be possible to prove it false, the hypothesis should be falsifiable.nothing can ever prove a theory right, it simple means it hasn't been proven wrong yet. This is why we start with the null hypothesis, if the results do not support the null hypothesis then this can be rejected and we can accept the alternative hypothesis with some degree of confidence.
Why is theory construction and hypothesis testing important in making psychology scientific
- in order to make predictions about events or behaviour, scientists need to create theories, these are a set of principles or laws that explain such events. Hey indicate a relationship between a set of variables. Theories can be created through induction and modified through the process of deduction. It should be possible to make clear and precise predictions based on the theory which are stated in the form of testable hypotheses. Empirical methods are then used to determine whether the hypothesis is supported or refuted by the evidence. If the results from the research challenge the hypothesis then the theory it came from needs to be modified.
How how paradigms and paradigm shifts a feature of science
-a paradigm is a shared set of assumptions about the subject matter of a discipline and the methods appropriate to its study. Argued that during a normal science phase there is a dominant paradigm which remains despite occasionally challenges from disconfirming evidence.
Even though challenging evidence will occur causing scientists to question the dominant paradigm. Kuhn argues that this will cause a scientific revolution and a paradigm shift where a new paradigm will replace the previous accepted assumption.
Why is probability and significance an important feature of science
Deciding to accept the null hypothesis (due to chance ) or the experimental/ alternative hypothesis - due to the variable.
There are a number of statistical tests available to psychologist.
What anagram can help us remember all the statistical tests.
Silly children spread petty minded unrelated Whiney rumours, related
Sign test- difference, unrelated, nominal
Chi x2- difference, related, nominal
Spearmans rho- correlation, related, ordinal
Pearsons r - correlation, related, interval
Man Whitney - difference, unrelated, ordinal
Unrelated t test -difference, unrelated, interval
Wilcoxon- difference, related, ordinal
Related t test- difference, related, interval
What is a parametric test
What Criteria must be met to carry one out
Pearsons r, unrelated and related t test are all parametric tests. They are more powerful and rebuts than other tests, meaning they may be able to detect significance within some data that non parametric tests cannot. Therefore wherever possible a psychologists will us a parametric test.
Data much be internal
Data drawn from a population that would be expected to be normally distributed for the variables being measured. If variables produce a skewed distribution then parametric tests cannot be used.
There should be a homogeneity of variance. So the sets of scores in each condition should have a similar dispersion or spread, the variance is the square of the standard deviation. The variance of condition one should not be more than four times greater the variance of other conditions.
How is the degrees of freedom calculated in the different tests
Chi2 - df= (number of rows -1) (number of columns -1)
Pearsons r -df= n-2
Unrelated t - df= (n1 + n2) -2
Related t test - df = n- 1.
What happens if psychologists pass the test at the level set
Results are said to be significant. This means that they have shown there is an acceptable probability that the results were not due to chance but due to the IV that was manipulated in the research.
Can say that your alternative hypothesis was supported and reject the null.
What errors can be made when deciding whether to accept or reject out hypothesis
Type 1- when we accept our alternative hypothesis when in fact it is false and we should of accepted the null hypothesis.
Type 2- when we reject the alternative hypothesis when it is in fact true. Accepted the null but should of rejected it.