research meathods y1 and 2 Flashcards
what are case studies
an in depth investigation, description and analysis of a single individual, group or institution or event. Usually an unusual individual e.g. Clive Wearing. or event e.g. events that lead to London riot 2011
-longnitudal usually and usually create qualititive data
evaluation of case studies
+ rich and detailed insights on unusual forms of behavior. Preferred to ‘superficial’ forms of that that may be collected from an experiment or questionnaire. They may generate hypothesis for future studies.
-such small sample sizes lead to generalization. Final report is also based on interpretation of researcher. Personal accounts from participants may be subjective to things such as memory decay, inaccurate.
what is content analysis?
type of observational research that enables the indirect study of behavior by examining communications that people produce e.g. texts emails words sentences phrases
what is coding?
initiate stage of content analysis that includes the communication to be studies is analysed by identifying each instance of the chose categories which may be e.g. phrases.
-quantitative data
what is thematic analysis?
qualitative data, form of content analysis.
inductive approach to analysis which involves identifying implicit or explicit ideas within the data, Themes will often emerge once data has been coded. (words that keep ‘cropping’ up as part of the communication is studied.
evaluation of coding analysis
+can get around many of the ethical issues normally associated with psychology research. Many of the materials they want to study may already exist within the public domain. These also benefit being high in external validity
-analysis is usually outside of context of communications due to communications being studies indirectly. Therefore there is a danger the researcher may use opinions to shape results or research content.
what is the experimental method?
manipulation of IV to measure the effect on DV, they may be lab, field, natural or quasi. Steps usually go: aim, hypothesis, deciding which hypothesis to use and doing the experiment.
whats a hypothesis?
clear, precise and testable statement that states the relationship between variables to be investigated. Stated at the outset of any study.
non directional hypothesis:
does not state the direction or difference between the relationship, example: people who drink coke differ in terms of talkativeness compared to those who dont drink coke.
directional hypothesis:
states the direction and difference between the relationship: people who drink coke become more talkative to people who dont.
what is the independent variable?
aspect of experimental situation which is changed or manipulated by the researcher or naturally so they can measure its affect on the DV
what is the dependent variable?
variable that is measured by the researcher. any affect on the dv should be caused by the change in the IV.
what is operationalisation?
clearly defining variables in terms of how they can be measured. Operationalised hypothesis example:after drinking 200ml of coke, Ps say more words in the next five mins than Ps who drink water. DIRECTIONAL OPERATIONALISED HYPOTHESIS
what are extraneous variables? (research issues)
any other variable other than the IV that affects the DV if it is not controlled.
what are confounding variables? (research issues)
kind of Ev where we cannot be sure if the affect is due to an EV or due to the IV. Example,they were chattier because of the energy drink or because they were excited to meet new people.
what are demand characteristics (research issues)
any cue from the researcher or situation that may be interpreted by Ps as revealing the purpose of an investigation. This may lead to Ps changing their behavior as a result of this.
what are investigator effects (research issues)
any unconscious or conscious effect of the investigators behavior on the research outcome. May include, design, study, or interaction with Ps.
what is randomization (research issues)
The use of chance methods to control for effects of bias when designing materials and deciding the order of experimental conditions
what is standarisation? (research issues)
Using the exact same formalized procedures and instructions for all Ps in a research study
what is experimental design?
way in which Ps are organised in relation to the experimental conditions
what are the three experimental designs?
Independent groups, repeated measures, matched pairs
what is independent group designs?
two separate groups experience two separate conditions.
what is repeated measures design?
all participants experience both conditions
what is matched pairs design?
Pairs of participants are first matched on some variables that may affect the dependent variable. Then one member is assigned to condition A and one to condition B
independent groups limits
-participants who occupy the groups are not the same in terms of participant variables. If there is a large mean difference between the groups this could be due to the participant variables rather than the IV. They may act as a confounding variable, reducing validity.
-less economical, twice as many Ps to provide equivalent data to that collected in repeated measures design.
TO DEAL WITH THIS: random allocation of participants in groups
repeated measure limits
repeating the tasks may be difficult if the order affects the result. Example, energy drink may still be in system when they then try the water.
TO DEAL WITH THIS: counterbalancing- half experience it in one order and half the other order
demand characteristics may also be a thing since they may have worked out the aims of the study while being subject to two conditions.
matched pairs limits
participant variables are still hard to avoid as matched pairs are not exactly the same.
Matching may be time consuming and expensive, particularly if a pre test is required, so this is less economical than other designs.
name four types of experiments
lab experiments, field experiments, natural experiments, quasi experiments
what is a lab experiment?
takes place in a controlled environment within which the researcher manipulates the IV and records the affects on the DV, whilst maintaining strict control on strenuous variables.
lab experiment: strength and weakness
+high control of extraneous variables, ensuring DV is only affected by IV (high internal validity). Replication is also therefore easier due to high levels of control.
-low external validity, not generalisable. Ps are usually aware they are being tested in this environment, so we may see demand characteristics.
what is a field experiment?
takes place in a natural setting within which the researcher manipulates the IV and records the affect on the DV.
field experiments strengths and weaknesses
+high mundane realism, natural environment, so more valid an authentic behavior. High external validity
-loss of control of CVS and EVS.
-important ethical issues, if Ps are unaware they are being studies they cannot give consent, invasion of privacy.
what are natural experiments?
the change in the IV is not brought about by the researcher, something else causes the IV to vary. The researcher records the effect on the DV they have decided. Example: before and after natural disaster./
natural experiments strengths and weaknesses
+Opportunities for research that otherwise may not be able to be taken due to practical and ethical issues.
+high external validity because they involve the study of the real world issues and problems as they happen, such as affects of natural disasters on stress levels.
-rare occurrence, and this may also limit the scope to generalizing the findings.
what are quasi experiments?
the IV is already an existing experience no one has manipulated, such as age or gender.
quasi experiments strengths and weaknesses
+often carries out under controlled conditions so therefore share strengths of lab experiences e.g. replication.
-cannot randomly allocate Ps to conditions so therefore there may be confounding variables.
what is a population?
group of people who are the focus of the researchers interest, from which as smaller sample is drawn
what is random sampling?
all members of the population have an equal chance to be picked.
first complete a list of all the members of the target population, then all the names on the list are assigned a number. The actual sample is selected through lottery form
what is systematic sampling?
when every nth number of a target population is selected. sampling frame is ordered and then the researcher picks the sample through, for example, every third person.
what is a stratified sample?
when the composition of the sample reflects the proportions of people in certain subgroups. example, 70% man U fans and 30% are Man City fans, 7 Man U fans will be selected while 2 Man City fans will be. Each of these would be randomly selected by their larger group- e.g. Man City fans choose the Man City participant.
what is an opportunity sample?
anyone who happens to willingly be available
what is a volunteer sample
the participants of the sample select their selves, for example through seeing an advert a researcher put up.
Random sample evaluation:
+potentially unbiased. CVs or EVS may be distributed equally between the groups, enhancing internal validity.
-difficult and time consuming. Complete list may be hard to obtain
systematic sample evaluation
+once system is selected, the researcher has now got no influence over who is chosen .
-time consuming
stratified sample evaluation
+produces a representative sample as it is designed to accurately represent the composition of the population. Means generalization may be possible
-identified strata cannot reflect all ways that people are different, so rep of the target population is not possible
opportunity sample evaluation
+less costly as list is not needed.
-cannot be generalized, as it is drawn to a specific area.
-researcher has complete control over selection so researcher bias is present
volunteer sampling evaluation
+less time consuming as the sample comes to them.
-volunteer bias may be a problem; asking for volunteers may attract a certain profile of person, e.g. one who want to please the researcher.
name four ethical issues
informed consent, deception, protection from harm, privacy and confidentiality
what are ethical issues?
arise when a conflict exists between rights of Ps and research goals
what is informed consent as an ethical issue?
making Ps aware of the aim of the research, procedures and their rights- including right to withdraw- and what their data will be used for. Ps should make a judgement whether they want to take part without being coerced or obliged.
This may be meaningless for the study as the P may know aims of study and behavior may not be natural
what is deception as an ethical issue?
deliberately misleading or withholding information from the participant about the investigation. It can be justified in situations if it does not cause undue stress.
what is protection from harm as an ethical issue?
Ps should not be placed in anymore risk as they would in their daily lives, and should be protected from physical and psychological harm. An important feature of this, for example, is Ps being told they can withdraw whenever they want.
what is privacy and as an ethical issue?
Ps have the right to control information about themselves. Confidentially should be protected
BPS code of conduct (ways of dealing with ethical issues)
the British psychological society has a code of ethics which includes a set of ethical guidelines
It instructs psychologists in the UK about what behavior is and is not acceptable . The code is built around the four principles of: respect, competence, responsibility and integrity.
dealing with informed consent (ways of dealing with ethical issues)
Ps should be given a consent form that included all relevant information needed. A parental signature is needed for under 16s.
dealing with deception and protection from harm (ways of dealing with ethical issues)
at the end of the study, the Ps should be given a full debrief at the end of the study and be made aware of the true aims of the investigation.
right to withhold data will be explained while researchers explain what there data is being used for.
Reassurance that behavior was typical or normal to the Ps if they are worried
dealing with confidentiality
if personal data is held, it must be protected. Other times the Ps data will not be recorded- anonymity. Example ‘HM’.
Ps will also be constantly reminded that their data will be protected an not shared.
what is a pilot study?
small scale version of an investigation that takes place before the real investigation is conducted. The aim is to check if procedures, measures etc work and for the researcher to see in they need to make and changes
what is a single blind study?
study where the researcher is aware but the Ps are not, as any info that might create expectations is therefore not known. Example, researcher is aware of who has the placebo pill and who does not, but Ps are not
what is a double blind study?
researcher and participants are both not aware of any info that might create expectations, example researchers influence of participant behavior.
control group?
purpose is comparison to the experimental group
naturalistic observation
watching and recording behavior in the setting it would normally occur.
controlled observation
watching and recording behavior in a structured environment ie one where variables are managed, such as a lab.
covert observation
Ps behavior is watched and recorded without their knowledge or consent. For this to be ethical it can be in a public setting as the behavior is happening anyway
overt observations
Ps behavior is watched and recorded with their consent
participant observation
researcher becomes a member of the whos behavior they are watching and recording
non participant observation
researcher remains outside the group they are watching and recording
Observations overall- evaluation
+give special insight into behavior
-observer bias
Naturalistic and controlled evaluation
+Nat- have high external validity, generalisable
-Nat- difficult to replicate
+C replication becomes easier as confounding variables are not a factor
-Low external validity
covert and overt observations evaluation
+Cov- demand characteristics are reduced so internal validity increases
-Cov-Ethical issues, even when in public
+O-ethically better
-O demand characteristics, low internal validity
participant and non participant observations
+Par-increased insight into lives of people, maybe meaning external validity can be increased
-Par-Danger that the researcher may become to close to the Ps and loose objectivity, as opinions now have influence
+Non- allow researcher to maintain an objective, less danger of social lifestyle
-Non-may loose valuable insight to the people they are studying that could be gained if they were involved
what is inter observer reliability?
to make observations more objective and less bias, more than one researcher should be involved in the investigation. Data from both observers should be collected and compared to check for consistency
-might do a piolet study before hand and analyze those results then see thier differences in interpretations before doing the main one
ways of recording data- unstructured observation
the researcher writes down everything they see.
rich in detail
may be important in small scale observations with few Ps
ways of recording data- structured observation
if there is too much going on in an observation it is necessary to simplify the target behaviors that are needed.
explain behavioral categories (observational design)
Target behavior or event is broken down into components that are observable and measurable (operationalisation) for example the target behavior affection may be broken down into hugging, kissing.
What are sampling methods (observational design)
unstructured observations can be impossible once the behaviors become complex, so structured observations must use systematic ways of sampling their observations. These ways are: event sampling and time sampling
what is event sampling?
counting number of times a particular behavior (event) occurs in a target individual or group. For example, event sampling of confidence in a lesson would mean counting the number of times a student puts their hand up
What is time sampling?
recording the behavior within a pre-established time frame. For example, in a football match we are only interested in one player so we note the players behavior every 30 seconds
correlations- strengths
+quick and economical to carry out, there is no need for controlled environments and no manipulation of variables is required, and less time consuming as data from others can me collected
-correlation does not mean causation. third variable problems
what is a self report technique?
any method in which a person is asked to state or explain their own feelings, opinions and behaviors and experiences related to the topic
explain a questionnaire as a self report technique
a set of written questions used to assess a persons thoughts and/or experiences.
It may be used as part of an experiment to assess the dependent variable, for example views on legislation for a drug
two styles of questions for questionnaires:
open and closed
what is an open question?
questions for which there is no fixed choice of response and respondents can answer in anyway they wish. Tend to produce qualitative data, but these wide range of responses may be difficult to analyse
what is a closed question?
offers a fixed number of responses determined by a question setter. for example rating of 1 through 10. This provides quantitative. Easy to analyse but may lack depth
three types of interviews:
structured(predetermined questions), unstructured (no set questions, works alot like a conversation. Interviewee is encouraged to expand answers) and semi structured (list of questions but follow up questions are also allowed
Questionnaires strengths
+can be produced without researcher being present
+cost effective
+can be distributed quickly, meaning data is produced quickly
+quick to analyse or deep in detail depending on open or closed questions
questionnaires limits:
-responses may not be truthful. May put themselves in a positive light. - social desirability bias.
structured interviews evaluation
+straight forward to replicated due to their format.
-limited data richment due to interviewers not being able to expand or divert from topics
unstructured interviews evaluation
+View of better insight into interviewee
-may lead to risk of interviewer bias.
-Drawing data may be tricky as there may therefore be a lot of irrelevant information recorded.
-social desirability may also occur
issues that should be considered when designing an interview:
-standardized interview schedule to reduce the contaminating effect of interviewer bias
-ethical issues, how their personal issues are confidential
issues that should be considered when designing questionnaires
emotive language or leading questions, over use of jargon (technical terms not everyone will know) or double barrel questions and double negatives
what is qualitative data?
data that is expressed in words and is non numerical, for example feelings expressed by Ps.
what is quantitative data
data that is expressed numerically and can be counted, for example number of words spoken by a person in a minute
what is primary data?
information obtained first hand by the researcher for the purposes of a research project. In psychology, such data is often gathered in an experiment, self report or observation
what is secondary data?
information that has already been collected by someone else and predates the current research project. in psychology, such data might include the work of other psychologists or government statistics.
qualitative and quantitative data evaluation:
QUALITATIVE +richer detail, much broader in scope and allows Ps to go deeper into thoughts and feelings.
+higher external validity due to the researcher therefore getting a better insight into Ps.
-Difficult to analyse, patterns an comparisons are tricky to identify
-as a consequence, this may lead to subjective analysis which can lead to biases
QUANTITATIVE: the opposite of the above
primary data evaluation
+authentic data
-requires time and effort for the researcher, for example planning a experiment takes time and resources when compared to secondary data that can be acquired and analysed in a matter of mins
secondary data evaluation
+inexpensive and easy to get, requires minimal effort.
-information might be, for example, untrustworthy due to it being out dated or incomplete, or it might not match the researchers wants or objectives. This challenges its validity
what is a normal distribution?
symmetrical spreads of frequency data that forms a bell shaped pattern. The mean, median and mode are all located at the highest point.
what is a skewed distribution?
a spread of frequency data that is not symmetrical, where the data clusters to one end
positive skew
type of frequency distribution where long tail is on the right side of the peak and most of the distribution is concentrated on the left
negative skew
type of frequency distribution in which the long tail is on the left side and distribution is concentrated on the right
what is a peer review ?
the assessment of scientific work by others who are specialists in the same field to ensure that any research for publication is of high quality
whats a meta analysis
a form of research that uses secondary data. Number of studies are identified that have the same aims and hypothesis’. It can give overall statistical measures of difference or relationship between variables of a number of studies
main aims of peer review
-allocate research funding (decide whether it deserves funding)
-to validate quality and relevance of research
-to suggest amendments and improvements
evaluation of peer review
+peer doing review remain anonymous throughout this process as this is more likely to produce an honest appraisal.
-however this could be sued to criticize rival researchers
-Publication bias- more ‘headline grabbing’ or lean towards more positivity.
what is reliability?
refers to how consistent a measuring device is, includes psychological tests or observations which asses behavior.
ways of assessing reliability:
test-retest, inter observer reliability
what is test- re test (assessing reliability)
a method of assessing reliability of a questionnaire or psychological test by assessing the same person on two different occasions. Shows to what extent the test produces the same results- consistent
-have to make sure there is a gap between both tests so recall is not a reason for good correlation
what is inter-observer reliability (assessing reliability)
the extent to which there is an agreement between two or more observers involved in observations of behavior. this is one between correlating the twos observations. Rule is that if the total number of agreements/total number of observant is >+.80 then it has high inter observer reliability. There is also sub categories of: inter rater reliability (content analysis) etc.
How can you improve the reliability of a questionnaire?
-test-retest method.
-comparing two sets of data should produce a correlation that exceeds +.80.
-questionnaire with low re test reliability should have questions deselected or rewritten.
-example swap open questions with closed questions, or confusing ones with straight forward ones
How can you improve the reliability of interviews
-use the same interviewer each time.
-If this is not possible, make sure all interviewers are trained so they do not ask leading or confusing questions.
-More easily avoidable if interviews are structured
How can you improve the reliability of observations
-Making sure behavioral categories have been properly operationalised, and that they are measurable
-If they are not operationalised, then the observers will have to perceive them in their own interpretation, leading to inconsistency.
-If reliability is low, observers need more training in using categories.
How can you improve the reliability of experiments
-procedures must be the same every time.
-standardized procedures
What is validity?
the extent to which an observed effect is genuine. Does it measure what it was supposed to measure? (internal) Can be generalized beyond the research setting in which is was found? (external)
What is ecological validity?
Type of external validity- extent to which findings from a research study can be generalized to other settings and situations
what is temporal validity?
extent to which findings from a research study can be generalized to other historical times and eras, form of external validity
what is face validity? (ways of assessing validity)
for of validity in which a measure is examined to determine whether it appears to measure what it is supposed to measure. Example, does an anxiety test look like it measures anxiety
what is concurrent validity? (ways of assessing validity)
the extent to which a psychological measure relates to an existing similar measure. Example, IQ test matched similar results to Stanford Binet test. Close agreement between two tests would mean high concurrent validity- indicated when correlation exceeds +.80
Improving validity of experiments
-control group can help assess whether changes to Dv were due to IV.
-standardized procedures to minimize participant reactivity and investigator effects on the validity of outcome
-Single blind or double biln activity, reducing investigator affects or/and demand characteristics
Improving validity of questionnaires
-lie scale to assess consistency or responses and control affects of social desirability.
-Ensuring all Ps that data will be kept anonymous
Improving validity of observations
-ensuing behavioral categories are specific
-covert observations can reduce ecological validity even further, as the P does not know they are being observed and it can be in a natural environment
Improving validity- qualitative research
-Depth of detail that reflects Ps reality better can ensure it has high external validity.
-monitor interpretative validity of their conclusions- extent to which researchers interpretation matches Ps.
-Triangulation- use of a number of different sources as evidence, example data compiled through interviews with friends, family personal diaries, observations etc
what is statistical testing?
used in psychology to determine whether a significant difference or correlation exists- (consequently, whether the null hypothesis should be rejected or retained)
what is the null hypothesis?
The null hypothesis states that there is no relationship between the two variables being studied. (H0)
The alternative hypothesis is..
The alternative hypothesis states that there is a relationship between the two variables being studied. It is alternative to the null hypothesis, and is trying to prove it wrong. (H1)
The statistical test determines which hypothesis is true to rejects or accepts the null hypothesis
-
what is probability?
measure of the likelihood that a particular event will occur where 0 indicates statistical impossibility and 1 indicates statistical certainty.
what is significance
statistical term that tells us how sure we are that a difference or correlation exists. A significant result means the researcher can reject the null hypothesis.
significance level is.. (definition)
the point at which the researcher can claim to have discovered a large enough difference or correlation within the data to claim an effect has been founding, meaning they can reject the null and accpet the alt.
level of significance in psychology (number)
0.05 (5%) or written p≤0.05
what does the p ≤ 0.05 mean?
probability that the observed effect occurred when there was no effect in the population is equal to or less than 5%. So even when a researcher claims to find a significant difference or correlation there is still up to 5% chance that this isn’t true for the target population from which the sample was drawn. Psychologists cannot be certain, so they have settled on a conventional level or probability that results may have been affected by chance.
calculated value
the result number of a calculated value
what is the critical value
to check for statistical significance, the calculate value must be compared to the critical value. Critical value is a number that tells us whether or not we can reject a null hypothesis- cut off point.
the rule of R
statistical tests with the letter R are those where the calculated value must be equal to or MORE than the critical value. Other tests the calculated value must be less than or equal to the critical value.
One tailed and two tailed tests (how the researcher knows what critical value to use)
one tailed test is hypothesis is directional and two tailed if it was non directional. (doubled for tt tests)
probability levels double for..(how the researcher knows what critical value to use)
two tailed tests as they are a more conservative prediction.
Number of Ps in the study (how the researcher knows what critical value to use)
-N value on the table. for some tests degrees of freedom are calculated instead
level if significance (how the researcher knows what critical value to use)
0.05 is standard
More stringent levels of significance
for example 0.01, might be used in studies such as drug trials that could not, for practical reasons, be repeated in the future. They will sometimes check the more stringent P if the critical value and calculated value have a large difference, as the lower the p value, the more statistically significant the result is.
type 1 error
the incorrect rejection of a true null hypothesis (false positive)
type 2 error
the failure to reject a false null hypothesis (false negative)
we are more likely to make a T1 error if the…
significance level is too lenient (too high) e.g. 0.1 or 10% rather than 5%
we are more likely to make a T2 error if the…
significance level is too stringent (too low) e.g. 0.01 or 1% rather than 5%. Psychologies prefer the 5% as it balances the chances of making a T1 or T2 error.
paradigm meaning (features of science)
a set of shared assumptions and agreed methods within a scientific discipline
paradigm shift meaning (features of science)
the result of a scientific revolution where there is a significant change in the dominant unifying theory within a scientific discipline, occurs when scientists start to disagree with the paradigm.Example, against Einsteins theory of relativity.
theory construction (features of science)
process of developing an explanation for the causes of behavior by systematically gathering evidence and then organizing it into a coherent account (theory)
hypothesis testing (features of science)
a key feature of a theory is that it should produce statements (hypothesis) which can then be tested, only in this way can a theory be falsified.
what is falsifiability? (features of science)
the principle that a theory is not scientific unless it admits the possibility of being proven false.
replicability meaning (features of science)
the extent to which scientific procedures findings can be replicated by other researchers. Important role for determining validity of a finding. It is important scientists report their investigations with as much precision as possible.
objectivity and empirical method (features of science)
scientists must keep objectivity in an experiment- all sources of personal bias are minimized so not to influence research process. This is a basis of an empirical method- scientific approaches that are based on the gathering of evidence through direct observation or speaking.
three factors needed to decide a statistical test:
1)whether a researcher is looking for difference or correlation
2)in case of difference- what experimental design is being used
3)The level of measurement
decision 1 (choosing a statistical test)
difference or correlation? Look at hypothesis- if the word association is used it is a correlation.
decision 2 (choosing a statistical test)
experimental design? Only if it is a test of difference.
Related: matched pairs and repeated measures.
Unrelated: independent groups
decision 3 (choosing a statistical test)
level of measurement.
nominal, ordinal and interval
nominal data
categories, example ‘yes’ or ‘no’ categories. discrete data as one item can in one category.
ordinal data
Ordered in some way, example 1 to 10 scale.
-does not have equal intervals between each unit
-lack precision as it is based on subjective opinion rather than objective measures. Idea of what 4 on the scale means may be subjective. Due to this they convert scores to ranks and use that in the statistical testing instead.
interval data
scales that use equal precisely defined size.
-accepted units of measurements such as stop watch, scales.
-used in parametric tests, as it is the most precise form of data in psychology.
Unrelated t test
unrelated design (test of difference), interval data
Mann- Whitney
unrelated design (test of difference), ordinal data
Chi-squared
appears twice in the table, one under unrelated design (test of difference), and one under test of association or correlation. Both use nominal data.
Related t test
related design (test of difference), interval data
Pearson’s r
test of association or correlation, interval data
Wilcoxon
in the middle of the table. Related design (test of difference), ordinal data
Spearmans rho
Test of association or correlation, ordinal data
Sign test
related design (test of difference), nominal data
order of statistical test table- letters
C S C
M W S
U R P
order on table of data types:
nominal, ordinal, interval
order on table of types of tests
Unrelated, related, (test of difference) I test of association
