Research Methods - Psychology OCR - H167 H567

Strengths of Open-Ended Questions

• Rich qualitative data is obtained as open questions allow the respondent to elaborate on their answer.

Four Types of Interview

• Structure • Semi-Structured • Clinical • Unstructured

What's an Independent Variable?

The independent variable is manipulated by the experimenter in order to determine the effect it has on the dependent variable

Observer Narratives

The observer may take notes during the session and then go back later to try to collect data and discern behavior patterns from these notes.

Non-participant Observation

The observer remains external to those being studied.

Tally Counts

The observer writes down when and how many times certain behaviors occurred.

Types of Sampling

• Random Sampling • Systematic/ Stratified Sampling • Opportunity Sampling

Validity

"A test is valid if it measures what it claims to measure"

Ways of Reducing Demand Characteristics

A double blind technique - Which is when participants do not know what condition they are in. they are given a false account of the experiment to prevent them seeking clues.

What's an Aim?

A general statement explaining why the study is being conducted.

Advantages of a Pilot Study

A pilot study can help the researcher spot any ambiguities (i.e. unusual things) or confusion in the information given to participants or problems with the task devised.

Standardised Procedures

All participants should be treated the same way, e.g. given the same amount of time to complete the task etc.

Control of Variables

As far as possible confounding variables must be controlled.

Informed Consent

Before deciding to take part in research participants should be told the details of the study so they can make an informed decision. Parental consent is needed before studying anyone under the age of 16.

Controlled Observations

Behavior is observed under controlled laboratory conditions (e.g. Bandura Bobo doll).

Disadvantage of the Range

Can be influenced by one score that is very different from all the others.

Internal Reliability

Consistency of results across items within a test

Content Analysis

Content analysis is a research tool used to indirectly observe the presence of certain words, images or concepts within the media (e.g. advertisements, books films etc.). For example, content analysis could be used to study sex-role stereotyping.

Correlation Coefficient

Correlation coefficient is a mathematical way to shoe how closely related two variables are. The closer the relationship to +1.0 the greater the positive correlation and the closer the relationship to -1.0 the greater the negative correlation.

Privacy

Data collected should be treated confidentially.

Audio or Video Recordings

Depending upon the type of behavior being observed, the researchers might also decide to make actual audio or video-taped recordings of each observation session.

What's a Hypothesis?

Explains what the study is designed to test. It is a proposition made so that it can be tested to see if its true.

Independent Group Design

Each participant is selected for one group / condition. Participants are selected at random, i.e. names out of a hat, tossing a coin etc.

Standardised Instructions

Every participant must be given the same instructions.

Experimenter Expectancy

Experiments expectations have an effect on the participant's performance

Participant/Covert Observations

Here the observer has direct contact with the group of people they are observing.

Operationalisation

How the DV is being measured, it needs to be clearly defined i.e. to measure performance in Short Term memory we could look at the number of words recalled.

Hawthorne Effect

If people are interested in something and get attention from the researches they show a more positive response as they try harder at tasks this means their results are often artificially high and can make conclusions inaccurate, the opposite effect may occur if participants are uninterested.

Natural Experiments

In natural experiments changes on a variable can be measured but the IV is not manipulated by the experimenter and participants have not been randomly selected / allocated to groups.

Event Sampling

Involves observing a behavior in a variety of different situations and settings.

Time Sampling

Involves taking samples as different intervals of time, which may be random or systematic

Disadvantage of the Mean

It can be misleading if there are extreme scores.

Advantages of the Matched pairs Design

It controls some of the individual differences

Disadvantage of the Median

It ignores most of the scores

Advantage of the Range

It is easy to calculate and takes into account extreme scores

Advantage of the Median

It is unaffected by a few extreme scores

Advantage of the Mode

It is unaffected by one or two extreme scores

Advantage of the Mean

It takes all scores into account

Disadvantage of the Mode

Less sensitive (too varying results)

Descriptive Statistics

Once an experiment is completed there is a set of raw data. Two things can be done with this data, there are measures of Central Tendency and Measures of Dispersion.

Right to Withdraw

Participants should be aware that they could leave during the study or withdraw their data at the end.

Deception

Participants should not be deceived. However if deception is used then participants must be debriefed afterwards. I.e. the real purpose of the study is explained.

Protection of Participants from Mental or Physical Harm

Participants should not suffer psychological harm during the study.

A Directional / One Tailed Hypothesis

Predicts a particular direction eg Higher.

Subjective

Something that is effected by personal feelings, prejudice and interpretations.

Disadvantages of a Pilot Study

Sometimes the task is too hard, and the researcher may get a floor effect, because none of the participants can score at all or can complete the task - all performances are low. The opposite effect is a ceiling effect, when the task is so easy that all achieve virtually full marks or top performances and are "hitting the ceiling".

Natural Observations

Spontaneous behavior is recorded in a natural setting.

A Null Hypothesis

States that there will be no effect / change.

A Non Directional / Two Tailed Hypothesis

States there will be a difference.

Participant Observation

The observer actually gets involved and joins in with the group being studied.

Structured (Formal) Interviews

Structured (formal) interviews are like a job interview. There is a fixed, predetermined set of questions that are put to every participant in the same order and in the same way. The interviewer stays within their role and maintains social distance from the interviewee.

What's the Mean?

The Mean is when all the results are added up and divided by how many there are.

What's the Median?

The Median is the middle score when the data is arranged in order.

What's the Mode?

The Mode is the most common score

What's an Dependent Variable?

The dependent variable is the aspect that will be measured.

Matched Pairs Design

The design uses two separate groups of people matched on a one to one basis on variables such as age, gender etc. this controls some individual differences.

External Reliability

The extent to which a measure varies from one use to another.

Investigator Effects

The researcher can also affect the results of their experiment. Their expectations, personal attributes etc can have an effect on participants behaviour.

Non-participant/Overt Observations

The researcher does not have direct contact with the people being observed.

How Can we Ensure Reliability?

The test-re-test method is when the same test is given to the participants on another occasion to check that scores are similar.

Clinical Interviews

These are used to assess peoples mental state., it is structured but open ended.

Structured Interviews

These contain fixed pre-determined questions and ways of replying ie yes, no, don't know. E.g. A market research interview

Field Experiments

These experiments are carried out in a natural environment / setting. Ie a school, train, office etc.

Unstructured Interviews

These may contain a topic area for discussion but there are no fixed questions or ways to answer.

Interviews

This can be a face-to-face situation where one person (interviewer) asks questions to another (the respondent).

Repeated Measures Design

This design uses the same participants in both conditions.

Whats Does Dispersion Indicate?

This indicates the extent to which scores cluster around the mid point.

Questionnaire Survey's

This involves asking participants about their attitudes, behaviour or intentions. The aim is to obtain information of a specific population of interest by administering the questionnaire to a sample of that population. They can be conducted by post, phone, Internet or even in the street.

Opportunity Sampling

This involves using who ever is available and willing to take part.

Systematic/ Stratified Sampling

This is a modified version of random sampling that may involve selecting every 10th person from a phone book or selecting every 5th house in a street.

Random Sampling

This is a sample in which every member of the target population has a chance of being selected

Pilot Study

This is a small scale study carried out before the main one to check procedures, design etc. it will also highlight any problems so adjustments can be made before the real experiment.

Quantitative Research

This is when the data measuring behaviour is numerical.

Semi Structured Interviews

This may be the most successful approach as there are some prepared questions and the interviewee can expand their answers.

Qualitative Research

This method involves extracting from the participants experience i.e. through interviews etc.

Naturalistic Observation

This method is designed to examine behaviour without experimenter interference. It is a research method that looks at how people or animals behave in a natural situation. The observer may be disclosed or undisclosed to those taking part.

Correlation Analysis Method

This method simply looks at the relationship between two variables. However even if a relationship between variables is discovered, it cannot be concluded that one caused the other because there could be other possible explanations. Correlations yield no cause and effect relationship

Participant Reactivity

This refers to a situation in which knowing that you are being studied affects behaviour. Regardless of the IV behaviour may change because they are being analysed. One example of this is demand characteristics.

Reliability

This refers to the extent to which the results are consistent. E.g. a persons IQ should be similar if not the same when tested again.

Internal Validity

This refers to whether the results are valid and can be linked to the manipulation of the IV. Results are internally valid if they have not been affected by confounding variables

External Validity

This refers to whether the results can be generalised beyond the experimental setting... - other settings (ecological validity) - other people (population validity) - over time (historical validity).

Objective

Unbiased or not influenced by private emotions, perceptions, or biases

Unstructured (Informal) Interviews

Unstructured (informal) interviews are like a casual conversation. There are no set questions and the participant is given the opportunity to raise whatever topics he/she feels are relevant and ask them in their own way. In this kind of interview much qualitative data is likely to be collected.

The Experimental Method

Use's manipulation and controlled testing to understand causal processes. Generally, one or more variables are manipulated to determine their effect on a dependent variable.

How do you Measure Dispersion?

Via the range (highest number minus the lowest number)

Research Design

When designing an experimental study, the researcher must have an aim and a hypothesis.

Open-Ended Questions

When the researcher does not restrict the range of available answers. This provides greater depth of qualitative information but answers are harder to analyse.

Closed Questions

When the researcher provides possible answers. This is usually used when factual information is required. It makes data easy to analyse but may lack realism due to forced choice of answers.

Steps of Designing a Questionnaire

• Aims - Make sure that all questions asked address the aims of the research. • Length - The longer the questionnaire the less likely people will complete it. • Pilot Study - Run a small scale practice study to ensure people understand the questions. • Question order - Easy questions first progressing to more difficult questions. • Terminology - There should be a minimum of technical jargon. • Presentation - Make sure it looks professional, include clear and concise instructions.

Advantages of Correlation Analysis

• Allows study of hypotheses that cannot be studied directly • Can obtain a large amount of data • Correlation studies can indicate trends which may lead to further experimental research • Problems of interpretation are reduced when no association is found.

Disadvantages of a Natural Experiment

• Replication is usually not possible • A cause and effect relationship is difficult to establish • The experimenter does not have total control over the situation.

Disadvantages of a Controlled Observations

• Awareness of being observed may affect participants' behaviour. • Demand characteristics. • Lower ecological validity. • Can't infer cause and effect although tentative conclusions may be drawn.

Advantages of a Field Experiment

• Behaviour of participants is often more natural. Similar to their normal behaviour • Can be replicated • High external validity - can relate results to real life situations.

Advantages of Naturalistic Observation

• Behaviour will be natural / realistic • Useful if other methods aren't appropriate. • Has high external validity • Results are full and rich with information

Disadvantages of the matched Pairs Design

• Can be difficult to accurately match pairs. • More participants are needed than with Repeated measures.

Advantages of a Lab Experiment

• Can be easily replicated • Supports theories / provides evidence • A cause and effect relationship can be established. • It is an original source of knowledge

Disadvantages of Correlation Analysis

• Cannot draw cause and effect conclusions • Interpretation of results can be difficult • Ethically - Researchers must be aware that their findings are not misinterpreted and become socially sensitive. (Intelligence and genetics)

Advantages of Questionnaire Survey's

• Cheap • Easy and simple to create and carry out • Quick • Varied sample • Can gain qualitative and quantitative information • Versatile - it can be used in a range of different situations • Closed questions are easy to analyse

Advantages of the Repeated Measures Design

• Controls all individual differences • Requires fewer participants than the other designs

Disadvantages of a Participant/Covert Observation

• Difficult to record data. • Difficult to be objective. • Difficult to replicate. • Observers may influence participants. • Ethical problems. • Cannot infer cause and effect.

Advantages of a Non-participant/Overt

• Easy to record data. • Easy to replicate. • Observers can't influence participants. • Ethical. • Can infer cause and effect.

Disadvantages of a Field Experiment

• Experimenter may have difficulty controlling every aspect of the situation • There are ethical issues. As informed consent cannot be obtained as they don't know they are taking part.

Disadvantages of Clinical Interviews

• Flexibility can lead to difficulty in replication

Advantages of Clinical Interviews

• Flexible • Reliable

Advantages of Semi Structured Interviews

• Flexible and Reliable • Sometimes easy to analyse and compare

Advantages of Unstructured Interviews

• Good validity - Interviewees will say what they think • Complex issues can be explored

Advantages of a Participant/Covert Observation

• High ecological validity if observation is covert. • Detailed and in-depth knowledge of the situation can be gained. • Qualitative data.

Ethical Issues in Psychological Research

• Informed Consent • Deception • Protection of Participants from Mental or Physical Harm • Right to Withdraw • Privacy

Advantages of a Natural Observation

• It allows researchers to study things that cannot be manipulated in a lab due to ethical concerns. E.g. prisons. • Ecologically Valid

Disadvantages of a Lab Experiment

• It is artificial • It is difficult to control all confounding variables • Investigator effects and demand characteristics are involved • There are ethical issues

Disadvantages of Questionnaire Survey's

• Lack of quality • Distorted data • Easy to write leading questions • Researcher effects - influence from researcher • Ambiguous / unclear questions

Disadvantages of a Structured Interview

• Less validity- The researcher cannot follow up on new / interesting lines of enquiry

Disadvantages of a Non-participant/Overt

• Low ecological validity if observation is Overt. • Demand Characteristics

Advantages of a Controlled Observations

• More control so observations are more accurate. • Easier to replicate than less controlled methods. • Usually avoids ethical problems of consent.

Ways to Ensure Internal Reliability

• No investigator effects • No demand characteristics • Use of standardised instructions • Use of a random sample

Disadvantages of Unstructured Interviews

• Not always reliable • It is difficult to analyse data

Two types of Observations

• Participant observation • Non-participant observation

Disadvantages of the Repeated Measures Design.

• Participants are likely to guess the aim / purpose of the study.

Advantages of a Natural Experiment

• Participants behaviour will be more natural • Data gathered is full and rich • Can be used whenever other methods are unethical or impractical

Participant Reactivity Examples

• Participants do their best to be good and try to guess what's expected of them. • Participants try to do the opposite of what's expected. • Participants may look out for 'tricks' so they can avoid being caught.

Disadvantages of Semi Structured Interviews

• Particular phrasing of questions could lead to low reliability • Very open ended questions / answer's are difficult to analyse

Disadvantages of a Natural Observation

• People may behave differently when they know they are being watched. - Demand Characteristics • People may try to behave in a certain way in order to conform with what they think the researchers expects to see - Biased Results? • Different observers may draw different conclusions from the same witnessed behaviour.

Disadvantages of Naturalistic Observation

• The researcher has no control • There are problems with reliability and replication • Some participants might change their behaviour if the know they are being observed • It can be hard to record information.

Advantages of the Independent Group Design

• There are no problems with order effects • No participants are lost between trials

Advantages of a Structured Interview

• There is less chance of deviating from the topic • It is easy to analyse • There is less risk of interviewer bias • Interviewees can be compared • Are reliable and replicable

Disadvantages of the Independent Group Design

• There may be significant individual differences to start with. • More participants are needed than within a repeated measures design

Strengths of Closed Questions

• They can provide large amounts of research data for relatively low costs. • The data can be quickly obtained as closed questions are easy to answer (usually just ticking a box). This means a large sample size can be obtained . • The questions are standardised. All respondents are asked exactly the same questions in the same order. This means a questionnaire can be replicated easily to check for reliability. Therefore, a second researcher can use the questionnaire to check that the results are consistent.

Weaknesses of Closed Questions

• They lack detail. Because the responses are fixed there is less scope for respondents to supply answers which reflect their true feelings on a topic.

Problems with Postal Questionnaires

• This is because some questionnaires may be lost in the post reducing the sample size. • The questionnaire may be completed by someone who is not a member of the research population. • Those with strong views on the questionnaire's subject are more likely to complete it than those with no interest in it.

Weaknesses of Open-Ended Questions

• Time consuming to collect the data. It takes longer for the respondent to complete open questions. This is a problem as a smaller sample size may be obtained. • Time consuming to analyze the data. It takes longer for the researcher to analyse qualitative data as they have to read the answers and try to put them into categories by coding, which is often subjective and difficult. • Not suitable for less educated respondents as open questions require superior writing skills and a better ability to express one's feelings verbally.