Extraneous variables Independent variable These are factors within

Extraneous variables Independent variable These are factors within an experiment that may interfere with the IV and DV. For example when testing the influence of age on memory anything that interferes with the effect of age on memory would be classed as an extraneous variable. For example, noise, reading ability of participants, light in the room etc. There are 2 types of extraneous variable: situational or participant. The IV is a variable (factor) that is directly manipulated by the researcher. For example when testing the influence of age on memory, age is the IV because it is what you are changing in the conditions of the experiment. Confounding variables If you don’t control extraneous variables they become confounding variables. This means that they will or have effected the results. Confound means damage, so they have damaged the results. Demand characteristics These are a type of situational variable in which the participant will change their behaviour to meet the needs of the study, for example they may have deciphered the true aim of the study and behaviour in a way they think the experimenter wants them to behave. This makes the results less valid. This can be made worse from Investigator effects as this is when the researcher may unitentially give the participant clues about how to behave e. g. nod when they give the correct answer. Situational variables These are a type of extraneous variable which may affect the cause and effect relationship between the IV and the DV. These are factors from the environment such as noise, light levels, temperature etc. An example may be ‘order effects’ this is a situational variable which means that the order in which participants did the experiment effected the outcome, for example if you are doing an eating test and you eat a bowl of chips and then are asked to eat a bowl of cereal second to see what fills you up better, by the time you eat the cereal you are already full from the chips so it’s a bad experiment! Standardised procedure The procedure of the study is kept the same across all conditions to reduce extraneous variables. Key words Independent variable – manipulated Dependent variable – measured Operationalise – Making the IV and DV specific to the study Dependent variable Extraneous variables – Factors which could influence the results The DV is a variable (factor) that is directly of a study measured by the researcher. For example when testing the influence of age on Confounding variables – An memory, the DV is the memory as that is extraneous variable that will effect what you are measuring. This may be done the results through the number of words recalled. Situational variable – An extraneous variable from the You must remember……. environment Participant variable - An ‘The IV is manipulated and the DV is extraneous variable from the measured!’ Extraneous variables may effect the results person of an experiment, confounding variables Demand characteristics – When will effect the results. the participants behave in a way they think is expected of them Participant Order variables effects – When the order These are factors associated with the participants interfere withthe results. This ofthat themay study effects could include IQ, personality type, memory ability, life experiences etc. The researcher only needs to consider what participant variables are likely to influence the findings of the study, e. g. driving ability is unlikely to effect the results of a memory tests but if the memory test is reading and remembering words then reading ability will influence the results. To counter act this the experimenter should screen all participants before and ensure they have the same or similar reading ages.

Hypothesis Null hypothesis Key words Every study has an experimental hypothesis and a null hypothesis. Hypothesis are different from a research aim because the aim shows the areas of interest for the study whereas the hypothesis states the predicted outcome of a study. PRECISE AND TESTABLE STATEMENT OF THE RELATIONSHIP BETWEEN TWO VARIABLES. A hypothesis is a statement about what is being tested and involves things that are measurable such as the IV and the DV. For example, there will be a significant difference between younger and older people in performance of a memory test. If the study is a correlation it does not look for a difference instead it looks for a ‘relationship’ and the word relationship should be used in the hypothesis. A null hypothesis is a prediction that the result will find no or little effect. They usually start with ‘There will be no significant difference between’ and end with ‘Any difference found will be due to chance’. Hypothesis – A predicted outcome of a study Null hypothesis – A prediction that there will be no difference or relationship between the IV and DV Experimental hypothesis – A hypothesis used in an experiment that suggests that there will be a relationship between the IV and the DV and the IV will cause an effect on the DV Directional hypothesis – An experimental hypothesis that suggests the direction of the results, e. g. an increase or decrease Non – directional hypothesis – An experimental hypothesis that suggests there will be a difference but doesn’t state what the difference will be. A non - directional (Two tailed) hypothesis This is a type of experimental hypothesis that clearly states there will be a difference but does not state what the difference will be. It still states that a difference or relationship will be found, but does not specify what that difference will be, for example: there will be a significant difference in the number of words recalled from psychology students then art students. Experimental (Alternative) hypothesis Just to add to confusion an experimental hypothesis is sometimes called an alternate or an alternative hypothesis. If it is an experiment (such as lab or field) it is an experimental hypothesis, It is called an alternate when it is a survey or interview (not an experiment). This is when we state the expected outcome of a study, where there will be a difference. This hypothesis is usually written based on a theory. A directional (One tailed) hypothesis This is a type of experimental hypothesis that clearly states the predicted direction of the experiment, for example there will be a significant increase in the number of words recalled from psychology students then art students. This is directional because of the word ‘increase’ which states the direction. Test yourself Create a directional and a non – directional hypothesis as well as a null hypothesis for a study with the aim of seeing the effect of hair colour on IQ score. Challenge – How would you test this? Can you design an experiment for it?

Stratified sampling Random sampling Truly random sampling only occurs when every member of the target population has an equal chance of being selected. Each individual is chosen entirely by chance and each member of the population has a known, but possibly non-equal, chance of being included in the sample. For example, putting names of every member of the target population into a hat and pulling a sample out (without looking). Strengths Should generate a representative sample because each member had equal chance of being selected, this makes it more generalisable. This helps to control participant variables giving the study more validity Weaknesses Participants who are selected may not agree to take part which would make the study less representative Not possible unless the target population is very small. Volunteer sampling Consist of those individuals who have consciously or unconsciously determined their own involvement in society, in other words they volunteer. For example, studies or passers by who become involved in field studies ie, in bystander intervention studies. Strengths – Involves minimal effort from the researcher Ethical as participants have fully consented to the research Weaknesses – Less representative because not everyone will see the advert or can respond Can lead to a bias sample as volunteers tend to be a type of person that want to take part. This may mean the findings are unreliable. Involves dividing the target population into important subcategories (or strata) and then selecting members of these subcategorise in the proportion that they occur in the target population. For example, if a target population consisted of 75% women and 25% men, a sample of 20 should include 15 women and 5 men. For example, suppose a farmer wishes to work out the average milk yield of each cow type in his herd which consists of Ayrshire, Friesian, Galloway and Jersey cows. He could divide up his herd into the four sub-groups and take samples from these. Strengths Ensures that the sample is totally representative of the target population. Weaknesses Time consuming and people can still refuse to take part, this can lead to a sample error which can lead to invalid conclusions being drawn. Opportunity sampling Simply involves selecting those subjects that are around available at the time, an effort may be made to not be biased in selecting particular types of subject. This may simply consist of choosing the first 20 students in your college canteen to fill in your questionnaire. For example, university psychologists may sample from their own students. Quickest and simplest technique Can lead to biased sample as not everyone will be available so not generalisable to the target population. Unreliable as replicating the study may not find the same results. Target population and sample A target population is the group of people that an investigation is concerned with or wishes the findings of the study to apply to. Because often it is impossible to study the whole target population we take a sample. For example I cant study every 15 year old in the UK but I can take a smaller sample of 40 different 15 year olds from different schools and study them. The sample chosen should represent the target population and thus allow the findings of the study to generalise to all the target population. A sample Is a selection of the target population that is directly studied in an investigation. Generalisability – This is the extent to which the results of a study represent the whole population not just the sample used.

Experimental and research designs Independent measures design Repeated measures design Matched pairs design When designing research, experimenters must chose a research design. This describes how participants are used within the study and in what conditions of the study are tested. An experimental design is the name given to a research design when used in an experiment. There are 3 types of experimental design. This involves splitting the participants in to groups and testing each group separately on one condition. Sometimes this is the only option, e. g. if testing the effect of age you would have to do 2 groups as one person cant be 2 ages at the same time. This is when the same participants complete all conditions of the study, for example all participants do both conditions; such as if an experiment was testing the effect of time of day on a memory task, all participants would sit the test at 9 am and then the same participants would sit the same exam at 5 pm and the researchers would compare the scores to see if time of day had an effect. This uses different people in each condition but tries to use similar participants to eliminate participant effects. This means when they recruit participants they would try and recruit similar participants, e. g. they match them all for age, IQ score, similar backgrounds etc. Strengths Weaknesses Controlling problems Independen t measures design There are no order effects as participants only take part in one condition so cannot get better through practice, or underperformance due to fatigue, or change their behaviour due to demand characteristics. It allows task variables to be controlled for example participants can be given the same word list in each condition so that this does not become a confounding variable. Any differences between conditions could be due to individual differences in participants, for example one group could do better on recall because they are motivated or more intelligent. Participants can be randomly allocated to each condition Repeated measures design Any differences between conditions are likely to be due to changes in the IV and not due to participant variables. Fewer participants need to be recruited, as they are used twice Order effects (e. g. practice effect, fatigue effect, recognising demand characteristics) as participants take part in all conditions. Order effects need to be controlled using counterbalancing or randomisation. Matched pairs design Fair comparisons can be made between the groups as they are equally matched – participant variables don’t interfere with the It is time-consuming to match all participants and not all characteristics can be equally matched. Key words Order Effects: An effect that can occur when a repeated measures design is employed. If the participants always complete one condition first, by the time they get to the secondition they may experience order effects, such as practice, boredom and fatigue. This could then affect their performance in the secondition. Demand characteristics - refers to an experimental artefact where participants form an interpretation of the experiment's purpose and unconsciously change their behaviour accordingly Experimenter variables - The experimenter effect is a term used to describe subtle cues or signals from an experimenter that affect the performance of participants in studies. The cues may be unconscious nonverbal cues, such as muscular tension or gestures. They may be vocal cues, such as tone of voice. Counterbalancing - Important control when using repeated measures as it reduces ‘carry over’ effects. Half PPs do condition A then B, Other half do condition B then A Participant variables – An extraneous variable from the participant that will interfere with the results.

Reliability Validity Reliability means to repeat and refers to the consistency of a result. If something is reliable it means if you repeated it you could expect to get the same results. If findings are reliable we can trust them and know they are not a one off. Reliability can be increased if a study has a standardised procedure. Validity refers to whether a study measures what it is supposed to measure. For example if a study is measuring obedience it is important that the measures actually measure obedience (this is known as internal validity). If the sample the study uses is representative of the target population and we can generalise the results then we would say the study has external validity. Key words Quantitative data – Numerical data, often collected through closed questions Qualitative data – Rich and detailed data often gathered through open questions. Researcher bias – when a researcher interprets the outcome of a study according their own view (this is also what subjective means) Triangulation – When more than one measure is taken for a behaviour to validate the findings. Objective – Not often to bias. The opposite of subjective. Quantitative data is often objective. Validity – Whether something measures what it is supposed to measure Reliable – If you replicate it you would get the same results. Quantitative measures This is numerical data. It is a scientific form of data. Often quantitative data is collected through closed questions from questionnaires and interviews. Quantitative methods are designed to gather facts and measure behaviour that can be applied to the target population so the data is generalisable and has external validity. This is more objective research method. Qualitative measures This is rich, in-depth and detailed data that often comes from open questions that allow participants to give extended responses. This type of data is more valid as it gives a much clearer picture of the thought, feelings and beliefs of the participants but is less reliable as it is harder to replicate and get the same answers. Primary / Secondary data Primary data is data the researcher has gathered for a specific piece of research – original research. Secondary data is data a researcher uses that was gathered previously for a different purpose. A problem with secondary data is that the original purpose of the first study may not quite suit the study that is reusing the data. But it can be cheaper to use existing data. Ecological validity Is the measure like a real life/natural situation? If an experiment is conducted in a laboratory it may be argued that it is NOT ecologically valid. If a self report has only used closed questions or Likert Scales it may be argued that it is NOT ecologically valid. If an observation is structured it can be argued that it is NOT ecologically valid

Ethics ETHICS are standards of conduct that distinguish between right and wrong, good and bad, justices and injustice. The primary aim of psychology must be to improve the quality of human life and to do this it is necessary to carry out research with human participants. Research psychologists have a duty to respect the rights and dignity of all participants. This means that they must follow certain moral principles and rules of conduct, which are designed to protect both participants and the reputation of psychology. The professional organisation that governs psychology in Britain is the British Psychological Society (BPS). They have produced a list of ethical guidelines that all practising psychologists must follow. An ethical issue is any situation that repeatedly gives rise to an ethical dilemma. Informed consent Deception Protection from harm Participants must give their consent to take part in research and this consent must be ‘informed’. This means that information must be made available on which to base a decision to participate or not. Participants should be told what they are letting themselves in for. Only then are they in a position to give informed consent. To study participants without consent would be ethically acceptable so long as what happens to the participants could just as likely happen to them in everyday life. E. g. observation in naturalistic setting such as bus queues. People in bus queues may be observed by anyone. In the case of young people under 16, consent should also be obtained from their parents. Deception means that information is withheld from participants; they are misled about the purpose of the study and what will happen during it. According to the BPS guideline ‘Intentional deception of the participants should be avoided whenever possible’. In particular deception is unacceptable if it leads to ‘discomfort anger or objections from the participants’ when the deception is revealed after the research has been complete’. The BPS ethical guideline states that ‘Investigators have a primary responsibility to protect participants from physical and mental harm during the investigation’. But, no investigation is risk free; the guiding principle is that risks should be no greater than the risks participants are exposed to in their normal lifestyles. Participants should be encouraged to contact the investigator after the research if they have any worries or concerns. The investigator has a responsibility to detect and remove any consequences of the research. A major problem with informed consent is the possibility that the researcher will ‘give the game away’ and thus influence participants’ behaviour. As a result, a case can be made for withholding information. Debriefing involves telling it all after the investigation is complete. Researchers should discuss the aims of the research with the participants, making sure they understand how they have contributed to meeting those aims. Any deception is disclosed, explained and justified. Attempts are made to undo any negative effects of the research. Codes of conduct for research state that all participants have a right to a debriefing session after the investigation has been completed. But, participants may be reluctant to express negative feelings during debriefing due to embarrassment or not wanting to upset the researcher But, the BPS accepts that sometimes deception is unavoidable. In such cases the researcher must: Make sure that alternative procedures which avoid deception are not available Consult with colleagues about how participants might be affected by the deception Reveal the deception to the participants immediately after the research has been completed. Right to withdraw Participants should have the right to withdraw from an investigation at any time. This is true even if they are being paid to take part. They should be told this at the start of the research. No attempt should be made to encourage or persuade them to remain. Confidentiality Protection from harm involves confidentiality. Participants may be asked personal questions. They must be told that there is no need to answer these questions and, if they do that their answers will be treated in confidence. That is, they should remain anonymous. Invasion of privacy can result in unease or distress. The BPS states that unless people have given their consent, they should only be observed in situations where they would expect to be observed by strangers. This limits observational research to public places.

Experiments Lab experiments Field experiments Experiments look for the effect that manipulated variables (independent variables, or IVs) have on measured variables (dependent variables, or DVs), i. e. causal effects. An experiment is an investigation in which a hypothesis is scientifically tested. In an experiment, an independent variable (the cause) is manipulated and the dependent variable (the effect) is measured; any extraneous variables are controlled. This type of experiment is conducted in a well-controlled environment (not necessarily a laboratory), where accurate measurements are possible. The researcher decides where the experiment will take place, at what time, with which participants, in what circumstances and using a standardized procedure. Participants are randomly allocated to each independent variable group. An example is Milgram’s study of obedience. Strength: It is easier to replicate (i. e. copy) a laboratory experiment. This is because a standardized procedure is used. Strength: They allow for precise control of extraneous and independent variables. This allows a cause and effect relationship to be established. Limitation: The artificiality of the setting may produce unnatural behaviour that does not reflect real life, i. e. low ecological validity. This means it would not be possible to generalize the findings to a real life setting. Limitation: Demand characteristics or experimenter effects may bias the results and become confounding variables. Field experiments are done in the everyday (i. e. real life) environment of the participants. The experimenter still manipulates the independent variable, but in a real-life setting (so cannot really control extraneous variables). Strength: Behaviour in a field experiment is more likely to reflect real life because of its natural setting, i. e. higher ecological validity than a lab experiment. Strength: There is less likelihood of demand characteristics affecting the results, as participants may not know they are being studied. This occurs when the study is covert. Limitation: There is less control over extraneous variables that might bias the results. This makes it difficult for another researcher to replicate the study in exactly the same way. Next steps Research the following studies and identify and justify which method of experiment they used and why this was the most appropriate: • Milgram (1963) • Zimbardo (1973) • Peterson and Peterson (1959) Natural experiments are conducted in the everyday (i. e. real life) environment of the participants, but here the experimenter has no control over the IV as it occurs naturally in real life. For example, Charlton’s St Helena study. Strength: Behaviour in a natural experiment is more likely to reflect real life because of its natural setting, i. e. very high ecological validity. Strength: There is less likelihood of demand characteristics affecting the results, as participants may not know they are being studied. Strength: Can be used in situations in which it would be ethically unacceptable to manipulate the independent variable, e. g. researching stress Limitation: They may be more expensive and time consuming than lab experiments. Limitation: There is no control over extraneous variables that might bias the results. This makes it difficult for another researcher to replicate the study in exactly the same way. Reliability and validity The more controlled an experiment (lab experiments) is the more reliable it is. This is because the standardised procedure means that it is more likely to be replicable. However, the more controlled an experiment is the more it lacks validity. This is because these extra controls often mean that participants know they are being studied and may change their behaviour accordingly. If they decipher the aim of the experiment they may demonstrate demand characteristics where they change their behaviour based on what they think is expected of them. Equally the artificial environment is likely to effect behaviour too as the setting isn't natural. This means the study will lack ecological validity – as it isn't like real life. Field and natural experiments have more validity because the setting is more natural for participants so behaviour is more normal, however because of this they cant have controls in place which means they lack reliability and we are

Ways to gather data Different methods can be used to investigate behaviour The nonexperimental method is still a scientific methods (as the information can be measured) and it can aim for objectivity (as it is fact based) Qualitative data is information that cannot be counted for example, about how you feel or think. It comes typically from asking open questions to which the answers are not limited by a set of choices or a scale, whereas closed questions generate quantitative data directly. Quantitative data is numerical information for example, about your age, how many hours do you work per week, how highly you rate a TV programme. It is the data that represents how much, how far and how long etc. there are of something ie, behaviour is measured in terms of numbers and quantities. Types of question Surveys Both questionnaires and interviews can contain open or closed questions. An open question is one that can be answered in any way the participant chooses. It yields qualitative data- data that consists of words that describe the participant’s views. A closed question limits the responses that can be made e. g. yes or no. It yields quantitative data- data that can be reduced to numbers or quantities. Surveys are a commonly used research method in social psychology. A survey can be thought of as an umbrella term for a number of different research designs including questionnaires and interviews. These methods investigate specific research questions by gathering self-report data. Self report data is data given about themselves STRENGTH WEAKNESS Represents the true complexities of human behaviour and gains access to thoughts and feelings that cannot be assessed using other methods More difficult to analyse so that conclusions are difficult to draw about the topic hypothesis More valid because the feelings, emotions and details of the situation are being taken into consideration Replication is more difficult because the results are detailed and descriptive so reliability may be lower Remember! Interviews tend to give data with more validity. However they can involve subjectivity and are hard to repeat. It is hard to test for reliability. Questionnaires are reliable and less likely to involve subjectivity. However, they tend to be less valid, as any open questions may be missed or answered briefly. Subjectivity / Objectivity Easier to analyse because the data is in number format, so graphs and analysis is more straight forward and conclusions can be drawn easily Reduces information about people to over-simplified statistics and so important information may be missed out, ie what and how people feel. Reliable because the results can be compared and replicated if Lower in ecological validity necessary and similar results are because the data is numerical likely. in format and doesn’t tell us about the descriptive situation, emotions or feelings of the research TIP! REMEMBER… Replicate = Reliable We are SUBJECTIVE when we consider something from our own perspective and we are OBJECTIVE when we see what is really there, unaltered by our own biases. Both questionnaires and interviews are subjective because the data analysis involves interpretation from the researcher. Questionnaires Set of questions that are designed to investigate a particular topic. Good way to get a large volume of data. Can ask open and closed questions. Questionnaires involve asking people what they think about a topic of interest. Questionnaires have to be designed carefully. They ask for personal data e. g. age, gender and background. Because questionnaires involve a written format there is no flexibility about the questions. Space can be left for the participants to write comments but otherwise set questions are answered. Questions are most likely to be closed and may make use of a Likert-type scale. You can use any format you like depending on your aim. More straightforward questions come first, in depth ones follow. Personal questions should come at the end as they take more time, so the respondent doesn’t get bored. Questionnaires should not be too long. A pilot survey should be carried out. This means… Questions should be set up in a way that allows a range of responses to be gathered to avoid response bias. Negatives should be avoided e. g. asking if someone is not a racist person. STRENGTHS the same questions are asked to all participants using the same standardised procedure this means that there is little variation in how people are asked the information, this means that data us realistic and valid and uninfluenced by the researcher. By using set procedures the questionnaires can be easily replicated as the same one can be used again therefore ensuring reliability. WEAKNESSES administering questionnaires can be difficult and this may mean other variables like location and others present could influence what the respondent will fill in and ultimately bias the results questionnaires often have restricted questions which means that the results could be invalid, closed questions may not offer enough options and open questions may restricted length, which leads to validity problems.

Interviews Evaluation of Interviews Researchers can ask closed questions during an interview but you can probe the participant to find out what lies behind superficial attitudes. Unstructured interviews are most likely to give qualitative data. Here the questions are open and the structure of the interview is flexible. There is a research question, which the interview is based around, but things are left quite unspecified to see what emerges during the interview. STRENGTHS - interviews enable a large amount of data to be collected which is descriptive and may give a better picture of what is going on in real life so are valid to what is being studied - interviews give access to information which is not available through direct observation, such as what individuals think and feel about certain topics which again makes it a more valid method. - interviews can gather a lot of information which can produce results which give insight into areas which may not have been thought of for example, if you are looking at people attitudes towards hard drug users you may not have thought of in your survey those who are ill and use prescribed drugs long-term. The structured interview involves a pre-set order of questions. This means there is little opportunity for the researcher to follow up areas of interest. WEAKNESSES - in interviews people often don’t know what they feel or do, and therefore are forced to rely on “social desirability”, meaning that they tend to answer a question in the way that seems most representative of “good” behaviour. This produces a SOCIAL DESIRABILITY a form of bias and reduces the reliability of the results. - in interviews the analysis of the information can be subjective especially if one person is carrying out the research, they may miss important information that others would pick up because of their personal opinions, ie RESEARCHER BIAS. Semi-structured interviews involve a schedule of questions that should be answered, but the researcher also has freedom to follow up on responses. Interviews mainly gather qualitative data, and so are used when indepth data is required. There are likely to be some quantitative data e. g. age, or yes/no questions. The more structured an interview is, the more likely it is to include quantitative data. Less structured interviews are more likely to generate qualitative data. Issues to consider- participants should see: The interview schedule (a set of questions and the time required). The chosen format for recording the interview. The full transcript of the interview (and agree with what has been recorded). Subjectivity and objectivity In all research, the researcher can cause bias. Social desirability, demand characteristics and response bias can all affect interviews. Researchers can cause bias by interpreting the results using their own views (subjectivity). Objectivity is when there is no bias affecting the results. Scientific studies must be objective. Correlations Evaluation of correlations This is a technique designed to look for relationships between variables. Can use an observation or questionnaire to gather data Looking for a relationship between co-variables. Plot a scatter gram to see. Positive correlation: each co variable increases together Negative correlation: as one increase another decreases • Allow us to investigate and test things that we may not be able to test otherwise, for example the correlation between smoking and lung cancer couldn’t be tested experimentally as it would be unethical to force participants to smoke to see if they get lung cancer! • However, we are unable to get a cause and effect relationship, as we cant be sure that the results are not down to other factors interfering. • Cheap and less time consuming

Observations Psychologists observe, they watch people’s behaviour and measure particular aspects in a way that is as precise as possible Usually have more than 1 observer to avoid bias Behaviour can be videoed. When observers interpret the data in the same way there is interobserver reliability. Researchers must first decide what behaviours are to be noted E. g. measure of aggression; kicking, punching etc. Naturalistic Conducted in an everyday environment where participant behave normally Controlled/structured Conducted in a lab – may be set up Overt Participants know they are being studied Covert Participants don’t know they are being studied Participant Observer becomes part of the group that they are observing Non-participant Observer takes a step back from the group Decisions over how and what time period the researchers will use Time sampling; record the child every 5 minutes for 25 minutes for example Rules need to be put in place before to ensure for standardisation. Behaviour may be analysed and broken down into behaviour categories such as; aggression, friendliness etc Controlled so is more reliable Structured Can be replicated Can see all behaviours Time and cost effective Manipulate a situation so a behaviour can be seen valid because they act normally If time sampling is used carefully with tallying and specific Naturalisti categories for behaviours and more than one observer it can be reliable and replicable c High ecological validity as in natural setting Participant No stranger present so act normally Data is valid are objective- stands back and observes Can record data easily Non Can use time sampling when tallying which cant really do participant if participating Exam tip It is important to specify which observation technique you are referring to in your answer. If you are criticising observations for ethical reasons, these may not apply to all observational techniques so you need to say which technique is associated with the ethical issue. lack validity as a constructed situation Demand characteristics most observations cant be replicated as the day/ time/ setting are important and one day will differ to the next Can be observer bias- experimenters see what the want to see Observer may be too involved to record the data- not objective and have to rely on memory Difficult to replicate as it is hard to find an observer who can also be a member of the group- grown up experimenter looking at teenage habits Your presence is likely to affect their behaviour- not natural Might miss behaviours if have to stand far away Might misinterpret behaviours if not involved Key terms • Inter – rater reliability: when more than one observer codes behaviour and their results are compared to check for agreement. • Observer bias: when an observer interprets the observed behaviour according to their own view.

Case studies A case study is a study of an individual or small group of people. It allows in depth and detailed data to be collected. Within this method, other methods can be used, such as interviews, observations and questionnaires. The case being studied occurs naturally, and the psychologist has no control over the situation. A case study is used, as the researcher cannot control any variables, he simply has to observe and study. Case studies allow a researcher to investigate a topic in far more detail than might be possible if they were trying to deal with a large number of research participants. A central tool which is used is the gathering of a case history. May include open ended interviews. Secondary data (data from schools etc) can be collected. May write a case study about a child having problems at school to try and solve these issues. STRENGTHS Data tends to be more valid, as it is in depth and focuses on real experiences in a real situation This is a valuable research method because it may be the only way to gather rich, detailed, qualitative information in context and with meaning for those concerned Sometimes it is the only way to be gathering data about a particular issue, for example, a child who has been deprived of parenting (you cannot set up an experiment to test such a thing) WEAKNESSES They lack generalisability because they are about one individual (or small group) only, so they are specific rather than in general The data gathered cannot be applied to any other case, individual or group, as the data is unique They are hard to replicate, so cannot be tested for reliability Data may be subjective, and therefore cannot be used to build a body of knowledge. Ratios are used to compare quantities. A ratio shows how much one thing compares to another. For example the ratio of 2 : 3 means for every 2 the first person gets the second person would get 3 parts. Standard form Decimals Standard form is a way of writing down large or small numbers without including all the digits for example… 10 x 10 x 10 10 to the power of 6 = 10 6 3200 written in standard form would be 3. 2 x 10 3 Decimal form refers to any number in the base-ten system Numbers are thought of as over 10 or in items of tenths So the fraction ½ written as a number ‘over’ 10 would be 5 over ten or 5/10 = 0. 05 Significant Figures Rounding numbers You are likely to be asked to round to 2 decimal places (2 dp) to do this look at the second number after the decimal place. Underline it and circle the number next to it, if that number is ‘ 5 or more’ you would round up. If not you would keep it the same. E. g 4. 5678 would round to 4. 57 because the 7 which is next to the 6 is ‘ 5 or more’. e. g. round 3268 to 1 sf the first significant figure is a 3, which represents 300, so we need to round to the nearest thousand. The number next to the 3 is a 2 which isnt ‘ 5 or more’ so we round down to 3000.

Percentages A percentage is a fraction of 100 To work out a percentage just work out the fraction and multiply by 100 Example: if 200 people were surveyed to find out how many thought practice, genetic make-up or good schooling were the reasons for something having a good memory A fourth option ‘do not know’ was added To find the percentage for each option divide the answer number by 200 and multiply by 100. Inferential statistics Descriptive stats are used to summarise raw data. The whole set of scores from a study. Before any analysis takes place. The mean, median and mode are types of average. The range gives a measure of the spread of a set of data. The table to the side shows how to calculate the mean, median, mode and range for a set of data. Strengths of this measure Weaknesses of this measure Adding the numbers up gives: 2 + 3 + 5 + 7 + 8 = 32 Makes use of all scores. This as good as it increased validity as all scores are included Takes into account extreme values…Why is this bad? The Median = middle To find the median, you need to put the values in order, then find the middle value. If there are two values in the middle then you find the mean of these two values. The numbers in order: 2 , 3 , (5) , 5 , 7 , 8 Easy to calculate Hard to with a small data calculate with set a large data set. Is easily affected by anomalies The Mode = most frequent The mode is the value which appears the most often in the data. It is possible to have more than one mode if there is more than one value which appears the most. The data values: 2, 2, 3, 5, 5, 7, 8 The Range To find the range, you first need to find the lowest and highest values in the data. The range is found by subtracting the The data values: 2, 2, 3, 5, 5, 7, 8 Measure The Mean = average Set A 2, 2, 3, 5, 5, 7, 8 To find the mean, you need to add up all the data, and then divide There are 7 values, so you divide this total by the number of values total by 7: 32 ÷ 7 = 4. 57. . . in the data. So the mean is 4. 57 The middle value is marked in brackets, and it is 5. So the median is 5 The values which appear most often are 2 and 5. They both appear more time than any of the other data values. So the modes are 2 and 5 The lowest value is 2 and the highest Easy with a small Doesn’t take data set into account all values Easy to calculate Impacted by and quick. extreme scores Shows us the

Normal distribution This is found when the mean, median and mode are very similar or the same. The further the scores are from the mean the less often they occur in a set of data. The graph will be symmetrical like below. If the mean, median and mode are not similar then a skewed distribution is produced. A normal distribution is found if the mean, median and mode for a set of data are very similar or exactly the same. When data are normally distributed 50% of the values are below the mean and 50% are above. The majority of the scores are equally spread close to the mean on either side of it. The further the scores are from the mean, the less often they occur in a set of data. Many mathematical statistical tests can only be carried out if data are normally distributed. Outliers Skewed distribution If the mean, median and mode are not similar then a skewed distribution is produced. Outliers can cause skewed distributions (the mean is very susceptible to outliers) A positive skewed distribution is caused by a high extreme set of scores, therefore a positive skew will contain more low scores than high scores. (The skew is caused by outlying positive scores) A negative distribution is caused by a low extreme of scores, and therefore will contain more high scores than low scores (the skew is caused by outlying low scores) Histograms in psychology refer to bar charts that show continuous data and thus don’t have gaps in between the bars. Scatter diagrams Only used for correlations Shows a relationship between two variables We add a line of best fit by drawing a line that has half the scores above it and half below. Bar charts Uses bars to describe categorical data As the data are discrete (not continuous) there are gaps between the bars Frequency Don’t forget ‘frequency’ refers to the ‘total’.