Scribd
Upload
11 views56 pages

Research Methods

The document discusses experimental methods in psychology, focusing on the manipulation of independent variables (IV) to observe their effects on dependent variables (DV). It outlines different experimental designs, including independent measures, repeated measures, and matched pairs, each with their strengths and weaknesses. Additionally, it emphasizes the importance of controlling for uncontrolled variables and demand characteristics to ensure valid results.

Uploaded by

aahanasolanki711
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF or read online on Scribd
11 views56 pages

Research Methods

The document discusses experimental methods in psychology, focusing on the manipulation of independent variables (IV) to observe their effects on dependent variables (DV). It outlines different experimental designs, including independent measures, repeated measures, and matched pairs, each with their strengths and weaknesses. Additionally, it emphasizes the importance of controlling for uncontrolled variables and demand characteristics to ensure valid results.

Uploaded by

aahanasolanki711
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF or read online on Scribd
[Aditi and AahanaSend fitery fc Research methods V4 Figure 1 Watch or imagine someone thinking really hard, perhaps trying to remember a name or work out the answer to a question, where do they focus their eyes? Ithas been suggested that in such situations we look upwards and to the let (Figure 1.1) Consider how you might test whether this is true. Would you wait for people to get confused and then look at what they do, or would you give them a puzzle to make them think? How would you decide where they are looking? What would you do to be sure that they aren't just looking around the room for clues? Being able to decide on the answers to questions such as these is the basis of designing experiments in psychology. 1.1 Experiments Anexperim to look for: investigates the way one variable, called th responsible forthe effect on another, the To test this the researcher fainiBHIBHES) the independent variable (IV) to produce two or more ‘levels’ or conditions, such as creating ‘bright or ‘dull’ lighting or selecting ‘early’ and ‘late’ in the day. The effect of thes jon the dependent variable (DV) is measured. For example, an IV of thedbrightness of lighting might affect attention, with people being better at paying, attention when the light is bright. How well people pay attention would be the DV. If there is a big difference in the DY between the conditions, the researcher would conclude that the IV has caused th the DY, ie. that the brightness of light affects attention (Figure |.2) lows researchers Th CAMBRIDGE INTERNATIONAL AS & A LEVEL PSYCHOLOGY: COURSEBOOK of such uncontrolled variables should re be minimised, e.g. by keeping all factors except lhe LV the same in each condition (or ‘level of the IV") The levels of the LV being compared may be two or more-eeperimenta conditions such as bright and dull artificial lights) or there may be one or more experimental conditions that are compared to asgontro -Condition(c.g. artificial light compared to daylight). The control condition is simply the absence of the ‘experimentalyariable, For example, in a comparison of the eflect of eating chocolate on paying attention, ve might compare either the effect of eating one bar or ‘wo bars (Iwo experimental conditions) or the effect of cating one bar to no chocolate at all (one experimental and one control condition). uncontrolled variable: a variable thateither acts | andomly affecting the DVin al levels of the | IV, or systematically, ie. on one level of the IV | (called a confounding variable) so can obscure | | | the effect of the IV, making the results difficult to interpret, the effects of which have not or cannot be limited or eliminated experimental condition: one or more of the situations in an experiment that represent different levels of the IV and are compared (or -ompared to a control condition) control condition: a level of the IV in an experiment from which the IV itself is absent. Itis compared to one or more experimental conditions ‘A researcher might conduct an experimentto test. | the effect of the independent variable af time of day on the dependent variable of happiness of students. They might always use lessons immediately after a break or lunch so the students had always eaten recently since this might affect happiness and act as an uncontrolled variable. This would be a comparison between two experimental conditions. 2 el Look at the Research methods in practice box. Can you suggest: © two different times of day to use as the of the independent variable how the dependent variable might be measured one other uncontrolled variable that it would | be important to control? Experimental design ‘The way that participants are allocated to different levels, of the IV iscalled the experimental design. They may participate in all, or only one, of the levels of the IV, The three experimental designs are: © ide pERGEHEHTERTTRES design * repeated:measuresidesign * matched pairsidesign experimental design: the way in which participants are allocated to levels of the IV independent measures design: an experimental design in which a different group of participants is used for each level of the IV (condition) Independent measures dedign In an independent measures design, a separate eroup of participants is used for each experimental condition or level of the IV. This means that the data for each level, the Vis ndepeadeat becuse is bot lated 2) other data: it has come from different people. Note that this is a different use of the word ‘independent’ from. ‘that im the ‘independent variable’. ‘If we wanted to know whether seeing people behaving aggressively on television hadvgmgetesnnelii¢etso © could (rather unethically) expose a group of young \ %, People to aggressive televiston and then wait for them to grow older. However, it would much quicker to compare ‘wo groups of adults, one group who had been allowed to watch aggressive TV as children and one group who hhad not been allowed to, This second example uses an independent measures design An independent measures design is good bec: participants only encounter th once They are therefore unlikely to notice or respond «© clues that micht tell them the aims of the experiment (demaiiehaPcteristies). One problem is that there mught be itividualidifferentes bet een participants in the groups for each level of the IV that could iifluence the findings. For example in a study on the eflect of noise on dreams, more people who normally remember their dreams well might end up in the ‘no noise’ group. If so, itcould look as though noise prevented dream recall when in fact it had little effect. This risk can be reduced by the fandantallacation of participants to diferent conditions. This spreads possible differences ‘between individuals across the levels of the IV. To randomly allocate participants, each person is given a ‘umber, and the numbers are then randomly divided into two groups. This can be done by giving each participant a number, putting numbers into a hat and drawing out two sets, o by using a random number generator (¢g. on a computer) to do the same thing. ‘This and other 9 independet ‘measures designs are presented in Table 1.1 se the demand characteristics: features of the experimental situation which give away the aims They can cause participants to ty to change 7 their behaviour, for example to match theie beliefs about what s supposed to happen, which reduces the validity of the study random allocation: a way to reduce the effect ©f confounding variables such as individual differences. Participants are put in each level of the IV such that each person has an equal chance cof being in any condition Repeated measures design 1 a eepeifG@lensuresdesign, the same group of people Participate in every level of the IV. To hel emember, think of the participants repeating’ their Performance under different conditions. For in a study looking at the effects of doodling on learning, we could count the number of words recalled in the same group of people when they did dood they did not le and when The main advantage of a repeated measures design is that each person acts as their awmsbaseline. Any differences between participants that could influence their performance and therefore the DV will alfect both levels of the IV in the same way. Individual differences are therefore unlikely to bias the findings. Imagine that in our experiment on doodling, one person was generally very quick to learn and another quite slow. In an independent measures design this might cause a problem if they were in diferent groups, but using a repeated measures design makes the differences between them less important, as both could show an improvement with doodling. Individual differences between participants that could affect the DV are called partiipanivariables. These variables, such as age, gender, personality or intelligence, could hide or exaggerate differences between levels of the IV confound or confuse the result. Such fasiablescan ‘be referred to as *eonfoundingwariables’. It is therefore to limit the powsibleefeets of these variables in order to be sure that any change in the DV is due to the IY. A repeated measures design does so by ‘@emnparing the same people in each condition ‘This and other strengths and weaknesses of re ‘measures designs are presented in Table 1.1 import the IV ed repeated measures design: an experimental design in which each participant performs in every level of the IV participant variables: individual differences between participants (such as age. personality and intelligence) that could affect their behaviour ina study that would hide or exaggerate differences between levels of the IV confounding variable: an uncontrolled variable that acts systematically on one level of the IV so could hide or exaggerate differences between levels and therefore ‘confound! or confuse the results making it difficult to understand the effect | ofthe IV on the DV CAMBRIDGE INTERNATIONAL AS & A LEVEL PSYCHOLOGY: COURSEBOOK ‘ ach individual participates in every level of the IV. hey will,perform the same ot two Or more limes. This can lead to a’ led an ondemetfect. Repeated performance can cause two different kinds of order effects. When participants\performance TAPS because they have encountered the task before this is a Practicereffect, This matters because participants who Were tested on a condition second would perform better than those who did it first. Alternatively, repetitions might make performance worse, perhaps if they Were bored or tired: a fatigneeffeet, In addition, the Participants see both levels of the IV and have more Opportunity to work out what is being tested, so are more likely to respond to demand chariteterstes ‘Ordereffeets can be solved in two ways: by randomisation or: Imagine an ‘experiment with two conditions: learning while listening jo music (M) and learning with no music (N). In qulinti’ participants are randomly allocated to Jo either condition M followed by N, or vice versa ‘As some will do each order, anyadvantage of d ‘ne of the conditions first will probably beevenedout {n the results, To be more certain that possible effects fre evened out, counterbalancing can be used. Here, {he group of participants is divided into two, so one palf will do M followed by N and the other half N followed by M. If on the second test there was a tiskeo participants accidentaly including items learned in the first test, this would be a problem for exactly half the participants in the ‘music’ condition, and exactly half in ihe ‘no music’ condition. Alternatively a different design ould be used order effects: practice and fatigue effects are the consequences of participating in a study more than once, for example in a repeated measures design. They cause changes in performance between conditions that are not due to the IV, so can obscure the effect on the DV practice effect: a situation where participants’ performance improves because they experience the experimental task more than once, for example due to familiarity or learning the task fatigue effect: a situation where participants’ performance declines because they experience the experimental task more than once, e.g. due to physical tiredness or boredom with the task randomisation: a way to overcome order effects in a repeated measures design. Each participant is allocated to perform in the different levels of the IV ina way that ensures they have an equal chance of participating n the diferent levels in any order counterbalancing: a way to overcome order effects ina repeated measures design. Each possible order of levels of the IV is performed by a diferent sub- group of participants. This can be described as an ABBA design, as hal the participants do condition Athen B, and half do B then A matched pairs design: an experimental design in which participants are arranged into pairs. Each pairs similar in ways that are important to the study and one member of each pair performs ina different level of the IV Matched pairs design The problems associated with both independent measures and repeated measures designs are overcome ina matched paiesdesignaFor each participant, another person i found who is similar in ways that are important to the experiment, such as age, gender, intelligence or personality, to make a pait. This matching is done on variables relevant to the study, 30 ina study on the effects of playing a violent computer | ‘game, participants might be matched on thei existing level of aggression. Identical twins make ideal matebed> they are both the same and are helo have had very (Figure 1.3), Figure 1.3: Identical twins are perfect participants for 8 ed pairs design : [independont measures Experimental design Strengths —_) Different participants are Repeated measures - used in each level | so there are no orde | Paricipants see only one level of the IV, | the effect of demand | characteristics | Random allocation to | | levels of the IV can reduce | | the effects of fidlividualy | | Silanes | \ | Weaknesses | hard to find orf participants Pantcipant variables are | | Participants unlikely to distort the effect | level of the IV Of the V, as each participant | the eff does all levels. charac Counterbalancing reduces | Part order effects. less likely to distort the effect of the IV than independent meas design as individual differences are matched. Uses fewer participants than independent measures or matched pairs so this is ‘good when participants are Different participants are Used in each level of the IV | so there are no order effects, are at risk, | Panicipant variables can | |e rere | ore pera nial | differences between | poricperts in ferent | levels of the IV. | More participants are | needed than in a repeated | measures design so the | study may bedessrethieal | | if participants are harmed | | and lessigecive if there | isa le because | panicpante ate hard to tnd. | Table 1.1: Strengths and weaknesses of experimental designs Different groups of participants are then used for ‘cach level of the IV, with one participant from each pair being in each level of the 1V. By using different Participants in eacl Order effects could distort The similarity between pairs the results istWfnitee! by the matching rocess, 30 the right As participants see the c Friabe measimmmaltask more than | chosen in advance for this to | once, they have @featen | prose 3° | to démand characteristics *vaiiability of matching pairs may be limited, making the sample size small altho some studies conducted o twins use very large numb of pairs) the matching of participants minimises the influence of individual differences. This and other strengths and Weaknesses of matched pair designs are preseated in Table 1.1 CAMBRIDGE INTERNATIONAL AS & A LEVEL PSYCHOLOGY: COURSEBOOK ? CAMBRIDGE INTERN, REN Ry Geeta oy ‘ a cOndUetedha: 0 look at the effect of violent computer gam (Figure 1.4) There were wo exparrenig Conditions (violent and non-violent), The dependent variable was the children’s subsequent Violent behaviour. The experimental design chosen was an independent measures design, with diferent children in each of the experimental conditions. fa repeated measures design had ‘been used, in which the same children played each type of game, there could be order effects. For example, aggression caused by playing the violent game could still aflect children in the non- violent game condition if they did this second. If this were the case, the problem could be reduced by using counterbalancing, Uap Hiledeaabanals n experiment However, the use of an independent measures design risks participant variables, such as the ‘original level of violence of each child, affecting the results. This could be reduced by either Using random allocation of participants to each condition or by using a matched pairs design. In | this case, children with similar aggression levels | would be put in the different conditions. To avoid demand characteristics, the children would | ideally be unaware that they are in an experiment, | pethaps by telling them that they are in a computer games competition. have Figure 1.4: Ave children more violet after they computer game than before? eee IATIONAL AS & A LEVEL PSYCHOLOGY: COURSEBOOK Look at the Research methods in practice box. Think about the following: +The study could have had another level of the IV that did not use a computer game but did use a computer, such as looking at non-violent pictures. Would this have been a control condition or another experimental condition? An order effect that could arise ifa repeated measures design was used for this study is that the children might get fed up with playing computer games by the second condition, Is this a practice effect ora fatigue effect? Suggest a participant variable, other than initia level of violence, that could affect the results of tis stud Types of experiment - Laboratory experiments Many experiments in psychology are conducted in artificial surroundings. Experiments conducted in this ‘way are called laboratargeexperiments; the participants are not in their usual ¢nyinonmentifor the behaviour they are performing. Theastificiaksituationmeans that the researcher can try to limit the effect of variables that might affect the DV. For example a laboratory experiment on the attention of schoolchildren in high and low light levels could be conducted. It might be investigated by testing the children on a computerised attention task conducted in a psychology room ina “university In this situation, the researcher could choose to control variables such as noise, that could also affect the children'seoncentration® laboratory experiment: a research method in which there is an IV, a DV and strict controls. it] Tooks for a causal relationship and is conducted | ina setting that is not in the usual environment for the participants with regard to the behaviour they are performing | a guecy 38 younger people eae WN. He tests them by company foe pressa button ay tet the emotionon a Each paricpont his is his W Quickly they they have face ofa screen. This is the Dv, its at the same dista Screen This is one aspect of the standard the procedure. Ina the participants time their reacti from the isation of Pilot study, Dr Singh had shown Pictures and used a stopwatch to ions himself, but he found he was Dot very eensistent in hisabimgto stop timing exactly when the participant responded, He therefore changed to the computerised system to Wpwrewelabsty. The pictures both younger and older valid test. The an that it would be possible for Dr Singh's coll Dr McArthur to replicate the study exactly People to ensure it was a \AESEERARTEESION mean league, ACTIVITY 1.3 Lock at the Research methods in practice box Write an operational definition for the independent variable, Write an operational definition for the dependent variable. Name and explain the experimental design bbeing used in this study. Suggest why the use of faces of a range of ages would have improved validity. Field experiments Returning to the idea atthe beginning of this section, about the effect of light levels on attention, the schoo! children could alternatively be tested by altering the ‘number of lights tuned on in their normal classroom. Light level would stil be the TV and the levels of the IV could be all the lights on’ and ‘half the lights on’. The DV of attention could then be measured by looking at the children’s scores on a topic test they were due to take ‘that day. This s still an experiment because it has an TV. and a DV (and there will still be some controls such as the amount of time they spend studying for the test). However, a 10 their normal ennan Evaluating field experiments Field experiments like laboratory experime: Fange of strengths and weaknesses in Table 1.2. Itis a li and than a laboratory experi cnt, although some control suill be used lower. However, validity ina field experim: ore be icpants ‘may ben are performing a task that seems normal, ina fannie environment, School students taken into a univer laboratory might concentrate really hard becawse they are nervous or interested, which might cover un amy uilferences between the clleren iagslevel conditions This means the findings from th ‘sould not tenner sings ashlee slasstoom, This isa problem of field experiments often have betes laboratory experiments (but not always) ‘eval validity than Another advantage ariss ifthe participants are unaware that they are in an experiment. If so, there may be fewer demand characteristics than there would be in a laboratory experiment. These are any features of the ‘experiment that give away theaimsand cause participants’ behaviour to change, for exampic to try to ‘make the ‘experiment work’. However, if participants are unaware that they have been in aguidymhey have not been given a choice about participating and may not have wanted to, field experiment: an investigation looking for 2 causal relationship in which an independent “Variable is manipulated and is expected to be ‘responsible for changes in the dependent variable. Ris conducted in the normal environment for the Pantcipants for the behaviour being investigated and some control of variables is possible | generalise: to apply the findings of a study more widely, e.g. to other settings and populations ‘ecological validity: the extent to which the findings (of research conducted in one situation would generalise to other situations. This is influenced by whether the situation (e.g. a laboratory) represents the real world effectively and whether the task relevant to real life (has mundane realism) validity (Causal relationships can be | Sood control of variables, raising 1 Research methods | As participants are in their normal stuation for the activity being studied, they are likely to behave naturally, making the results Getermined as only the IV should be | "*Pesentatis | affecting the Dv. | Standardised procedures raise reliability and allow replication. Weaknesses | The artificial situation could make participants’ behaviour lowering ecological validity. Participants could respond to their behaviour. Table 1.2: Strengths and weaknesses of experimental methods ‘[Link] team is deciding how to test the effect, cof watching television on children’ | aaa cc y wll ial behaviour by observing how often the children help a doll that cries. They are consideri }ods, both using an independent measures design. One is a | laboratory experiment, in which each child is ether shown a cartoon in which the characters are very helpful or a cartoon with characters who behave in a neutral way. Each child is then given a doll for 10 minutes and observed to count how many times they help the aying doll. Alternatively, the children could be observed in a field experiment. In this study, half of the children's parents would show their child the pro-social cartoon on the family demand characteristics [Link]. | !f participants are unaware that they are in a study, the problem of demand characteristics is less than in laboratory | | experiments. | Control of variables is harder than in | | laboratory experiments, lowering refabilty | | and making replication difficult. | The researcher will be less sure that changes in the DV have been caused by changes in the IV than in a laboratory experiment. Participants may be unaware that they are in a study, raising ethiesl Sues. television. The other half would show their child the neutral cartoon. Each child would be videoed ‘with doll for 10 minutes and the researchers would view the video to count how many times the child helped the crying doll. The field experiment would have more ecological validity than the laboratory ‘experiment because the children would be ina familiar environment, in which they would normally play, so would be less likely to be nervous than if they were in a laboratory. For this reason, the results of this study may generalise to other situations, such as performing pro-social behaviours in school ‘or at the park. However, a disadvantage of the field ‘experiment, is that there may be distractions in the home that prevent them from paying attention to the cartoon. CAMBRIDGE INTERNATIONAL AS & A LEVEL F | Want the Research methods in practice box | Whi you identify? independent variable of the following ca * Dependent variable, Does the field experiment have a contro! | condition? Can you suggest a variable that it would be important to controf? What effect mi ight demand characteris in this study? ics have What other details would you need to be able to replicate the laboratory experiment? Hypotheses in experimental studies At the very beginning of this chapter, we considered 4 study that Dr Huang was conducting, with the aim to investigate whether mind maps or revision apps are more effective at helping students todearm(Figure 1.5) Dr Huang would also need to write a hypothesis based on this aim. A hypothesis is as a nt. It Provides a little more detail about the variables being investigated than the aim. Importantly, a hypothesis should b at is it should be possible for it to be shown to be wrong. This is essential, otherwise an alternative hypothesis Would always be accepted! At A2 you may learn about the psyehodynaimiie®ipprolich and the work of a famous carly psychologist. Aspects of his work were cantemtiousypartly because some of his ideas were nonefalsifiables such as his ideas about ‘our behaviour being the consequence of unconscious motives. It has been argued that one problem with unconscious miG8V@"s that they are impossible to test, in anebj eve is a basic example: a young woman is slightly rude to a young man. Freud could ‘use two different explanations. One is that the young ‘woman unconsciously hates the young man, and she is avoiding being as rude as she unconsciously wants to be. The other is that the young woman unconsciously Joves the young man, and she is avoiding showing her unconscious desir so is behaving in a contrary way. The problem here is that Freud cannot be wrong-either YCHOLOGY: ‘OURSEBOOK she feels unconsciously fit with his explanations The Positive or negative-both would in hypothe ina study (sometimes called the can be wnitten in several dife ‘ways They differ in terms of the nature of thoprediatign they make about the results of an investigation hypothesis (plural hypotheses): a testable | Statement based on the aims of an investi ation | alternative hypothesis: the testable statement Predicting a difference in the DV between levels | ofthe independent variable in an experiment {or a relationship between variables in a correlation) Figure 1.5: Revision apps such as Quizlet and Gojmo ‘may, or may not, help students to learn Hypotheses should have o For example, in the hypothesis “Students using revision apps will learn better than students using mind maps’, “> we do not know how *betterearnifie’ will be measured, or which apps are being used because the variables are not operationalised. To be complete, the hypothesis needs more detail, such as ‘Students using the Gojimo revision app will gain better test marks than students using mind maps Non-directional hypotheses An edicts that there will be an effect, but not the direction of that effect (Figure 1.6). In an experiment, this means that the hypothesis suggests thatthe IV will change the DV but not whether the effect will be an increase ora decrease. This type of hypothesis is chosen ifthe effect of the variable is being tested for the first time, so there are no previous results to suggest what the results might be. For example, Dr Huang's hypothesis could be: There is a difference between the effeetivenesof mind maps and revision apps in helping students to learn. Note that itis predicting a difference, but not which condition will be better at helping with learning, Figure 1.6: Unlike with a one-tailed hypothofish, you can't see which way a two-tailed hypothofish will swim Directional hypotheses ‘When most previous research or other evidence suggests the nature of ‘direction’ of an effect we can use a ‘cdirectionatonestatled)strypothesiseln an experiment, this ‘means saying which condition will be ‘best’, i. produce the ‘highest’ scores (and in a. whether there will be apposi ion). Returning to Dr Huang’s study, there might be evidence that revision apps are better than mind maps, perhaps ‘because they are more ‘agtivesand being actively engaged helps memory. This is a directional prediction so the hypothesis might be: Students using revision apps will learn better than students using mind maps. Note that the opposite fpredictio®could also be expressed as a directional hypothesis. This would be: Students using ‘mind maps will lear better than students using revision ‘apps. We might make this prediction if we believed that writing a mind map yourself was more effective than Just re-using ready-made materials on revision apps. 1 Research methods Null hypotheses The alternative hypothesis isan alternative tothe ll) ‘wypathesis In an experiment, the null hypothesis states that any difference in the DV between levels ofthe IV 48 so small that tis lkely to have arisen by chance. The concept of chance becomes important iningerentiah “Satistics’. These are mathematical processes used to find ‘out how likely it is that the patternofiresults (a diference or a correlation) found from the sample of participants could have arisen by chance. This information helps us to infer whether the findings would apply tothe wider population, When you read about the corestudies, the most important results are the ‘significant’ ones. This word has a particular meaning, that there isa ‘mathematically significant probability thatthe pattern in the results could not have arisen by chance. So, when inferential statistics are used, a ‘significant’ pattern enables the researchers to reject their mull hypothesis and accept their altemativeliypothesi. If, on the other hand, the patter inthe results is non-significant, the researchers ‘must accept their null hypothesis: that te difference or correlation could have arisen byehance ‘There‘are two ways to write a null hypothesis for an experiment. In the case of Dr Huang’ stud, the null hypothesis could be writen either as: There wll be no difference between the effectiveness of mind maps and revision apps in helping students to learn or Any difference in effectiveness of mind maps and revision apps in helping students to learn is due to chance. statement predicting -_ i valve related to another, for example that there willbe a difference in the DV between levels of the IV in an experiment (or that there will be a relationship between the measured variables in a correlation) directional (6n6:t8i1ed) Fypothesis: statements predicting the direction of a relationship between variables, for example in an experiment whether the levels of the IV will produce an increase or a decrease in the DV (or in a correlation whether an increase in one variable will be linked to an increase or a decrease in another variable) a testablesstatement saying shat | any difference or correlation in the resultsisdue | | to chance, that is, that no pattern in the esults— | | has arisen because of the variables being studied see \ 1s > CAMBRIDGE INTERNATIONAL AS & A LEVEL PSYCHOLOGY: Ethics in experiments The role of;ethiesin psychology is discussed in Section 110. Hi e we will briefly consider ethics in experiments (Figure 1.7), participant in a laboratory experiment 'slikely to know that they are participating in a study and can readily be asked for theirinformed consent, However, forthe sake of yalidity it may be necessary to deceive them to avoid them working out the aim of the study and altering their behaviour, i. to reduce demand “Scharacteristi€s There is therefore a need to balance ‘good ethics and good science, such as high validity. Infield “experiments, in contrast, it may not be possible to gain ‘consent as the participants may be unaware that they are even in a study. This isamethicabproblem because Participants should have the right to know what they are entering into and to agree to participate or not. They should also have th ‘hich they cannot oi they do notes ee ey are in a study. In order to achieve a balance between good science and good ethics, itis therefore very important that participants in a field experiment are effectivelyprotestedfromharm, In experiments, whether they are in a laboratory or the field experimenters may sometimes need to actively lie to or mislead participants. This is called deception, and may be necessary to reduce the risk of demand characteristics affecting participants. Deception should be avoided and, where possible, the reality of the study should be explained to the participants afterwards in a debrief. This is often much harder to do in a field experiment than a laboratory experiment and it may even be impossible. The use of deception should, however, be avoided if possible informed:consent: knowing enough about 2 study to decide whether you want to agree to participate srightstonwithdraw®a participant should know they can remove themselves, and their data, from a study at any time protectionifromharmeparticipants should not be exposed to any greater physical or than they would expect in their day-to-day life ae Figure 1.7: Researchers mus {900d ethics and good scien: leve a balanc In all experimentsaprivacy andeonfidentiality ar important. Privacy can be respected in laboratory experiments because the tests or questions used arepres ‘Planted "ete natural setings ofa feld experiment, however, thee is a risk of invading a participant's ot pethaps discovering something that they wouldn't want the researcher to know so researchers must be careful of this. Confidentiality can be respected in all experiments by keeping the Participants’ data seeurtandeanonymous, altho if the participants are unaware that data has be collected, as in afield experiment, its important to ensure that they cannot be individually identified, for example by their place of work. articipants should not be deliberately misinformed (lied to) about the aim or procedure of the study. If this is unavoidable, the study should be planned to minimise the risk cf distress) and participants should be thoroughly debriefed. It may be done to reduce the effects ‘of demand characteristics but should be avoided ‘privacy: participants’ emotions and physical space should not be invaded, for example they should not be observed in situations or places. where they would not expect to be seen Participants’ results and personal information should be kept safely and not released to anyone outside the study A psychology department ethical committee is looking at aresearchypraposalifora study about the effect of on a therapy designed to help people See verons only plan to ask for consent about part of procedure they will use listening to an imagery-based relaxation recording - and not their full aim. Although they intend to tell ll the participants they will have pulse monitors and be in darkened rooms, they also intend to use deception. They will deceive the participants about which level of the independent variable they will be allocated to. Rather than telling them what will eally happen ~ their pulse rate should fall the researchers will give them false information by teling them that some people see disturbing flashing lights. When the participants ate given the limited information atthe start of the study they will also be told that they can leave at any time, thus giving them the right 10 withdraw. The instructions on the recording tell the Participants to imagine relaxing, intimate thoughts. ‘The participants will also be reassured that they wil ‘not be asked about these thoughts, which ensures their privacy is protected. When the participants join the study, each will be given @ number, which will be used to identify their data so that their names do ot have to be used, ensuring their confidentiality Look at the Research methods in practice box. Which of the following can you identify? ‘+ The type of experiment being planned + The experimental condition of the independent variable + The dependent variable + The experimental design Can you suggest one way in which participants would not be protected from harm in this study? Suggest why participants may want to withdraw. Why might it be necessary for the researchers to deceive the participants? What is wrong with the hypothesis ‘an imagery based recording will help people to relax’? 1 Research methods Applying your knowledge of experiments to Movel’research situations You should be able to recognise experiments (including the IV and DV, and be able to them, ie. define them in detail) to decide whether an experiment is a laboratory or field experiment and to evaluate them, cg. in terms of controls, standardisation, ethics, and reliability and validity In addition, you should be able to plan an experiment; choosing the type of experiment (laboratory or field), deciding on and operationalising an IV and a DV, Questions Barry and Anouk are deciding how to test whether gender affects artistic ability. Barry suggests doing a study in the psychology department where they ask students to come in fora study about memory in which they must redraw a complicated image. Barry and ‘Anouk can then see how well they do it. Anouk thinks it would be better to,persuadethe art teacher to use an art class and set a lesson where students have to copy the same complicated image. 11 Explain the type of experiment that is being suggested: a by Barry bby Anouk 2 The independent variable is the same in Barry's and Anouk’s studies, as is the dependent variable. a Describe the independent variable (IV), b Describe the dependent variable (DY). 3 Explain one ethical issue that is clear from the ‘sprocedurethat Barry suggested, 4 Suggest one other ethical issue and how they could, avoid problems with this issue. AM NTERNATIONA 12 Self-reports 1n a sett the partic ‘anformavon about themselves directly. This different trom experimental tests oo researcher finds the d Pant gives the researcher ervations where the 4 out from the participant. There are tWo Main ways to conduct a self-report: uain ‘eedtuestionnaire or amintgrmew. Both types of sel allow the researcher to ask the participa Questionnaires Ina questionnaire, the questions are presented to the Participant in written form, The two techniques for doing this are as {oman peng! exercise or gnline. ‘There are several different types of question. The two ‘most important question formats a Fe closed questions, hich have a fixed set of possible responses, and ape ‘Wostionsemhich ask for descriptive answers in the Paruicipant’s own words (Closed questions can take the form of simple choices, such as those asking for yevno answers or items from a list, Other forms of closed questions include rating) ‘ealestwhere a number is chosen, eg. between O and 5) ‘and IRGFP wes, which ask the respondent to say how much they agree with a statement such as ‘Obesity is not Important’ or "Exercise is a necessity’ using the choices “strongly agree / agree / neither agree nor disagree / disagree / strongly disagree’. Some examples of closed questions are as follows: + What is your gender: boy or girl? ‘= How do you travel to school? walk / bicycle / bus / train / car + Indicate which animal(s) scare you: dog, spider, cat, rat, fish, rabbit, bird. [You may tick as many as you like} ‘+ How much do you like psychology on a scale of (0-42 (0 = not at all, 4= very much) ‘ealfmepontes research method, such as a ‘questionnaire or interview, which obtains data by asking participants to provide information about themselves ‘questionnaire selfepor research method | | that uses written questions through a ‘paper and pencil’ or online technique | > AS & A LEVEL PSYCHOLOGY: COURSEBOOK eSlonmctqUBEHERE a question format in questionnaires, interviews or test items that produces quantitative data, They have only | a few, stated alternative responses and nc | | opportunity to expand on answers ‘OPEMETUBELIBAE: a question format in questionnaires, interviews or test items that Produces qualitative data, Participants give full and detailed answers in their own words, that is, ho categories or choices are given Open questions prompt the respondent to give detailed answers, which may be quite long. They contain more depth than the answers to closed questions and are more likely to be able to explore the reasons behind behaviours, emotions or reasoning. They typically ask “Why...? or simply ‘Describe... Some examples of open questions areas follows + Why-do you believe is important to help people wh sulle fromyphobias? Describe your views on the use of social media sites with regard to encouraging helping behaviour, Explain how you would respond if you were told to hurt another person, Evaluating questionnaires Questionnaires, which often use manly closed questions are faitlyeagytoanalye as they produce a total of each category of answers making it simple to summarise the findings Iis also possible to work out averages, which can belp to describe the pattems in the results Where _CAMBRIDGE INTERNATIONAL AS & A LEVEL PSYCHOLOGY: COURSEBOOK questions on what they said. A colleague, Or McArthur suggests that both methods have limitations. Interpreting the responses to the unstructured interview might lead to very subjective data, especially as Dr Singh already believes the finding that the new shopping centre layout encourages helpfulness. Although the data from the closed questions in the structured interview might produce more objective measures, this would limit opportunities for asking participants to expand on their answers. The colleague suggests that a semi-structured interview might be better. Xena Look at the previous Research methods in practice box. Why is the first of Dr Singh's suggestions a structured interview? ‘Why would the data from these questions be more objective? Why is the second plan an unstructured interview? What isthe problem with subjective interpretations ofthe participants’ responses, in the unstructured interview? Suggest why a semi-structured interview would be better in this case Questions 5 Shareen and Judith are investigating people's phobias. They have decided to use self-reports, Shareen is suggesting using a questionnaire and Judith wants to interview people instead @ Suggest one closed question and one open {question that Shareen could use. b Suggest one reason why Judith might want to. conduct an unstructured interview © De cribe one ethical problem that might arise in either Shareen's or Judith’s version of the study. 1.3 Case studies A casesstudyis a detailed investigation of a single instance, usually just one person, although it could bea single family or institution. The data collected is detailed and in-depth, and may be obtained using a variety of different techmiqueseFor example, in one case study, a participant may be interviewed, observed, given tests and asked to fill in questionnaires. Case studies are particularly useful for looking atwranescases where a detailed description is useful, and for following developmentak:changes where the progress of a child, or a person with a disorder, can be tracked through their improvement or decline. Case studies are therefore sometimes linked to therapy but itis important to | remember that when the case study as a research method | is being discussed, the therapeuticpurpose is not the ‘main aim. In addition to using different methods or techniques ‘within a case study, different sources ofjevidence may be used. These could include interviewing the participant themselves, their relatives or other people such as. colleagues to obtain a wide range of information about the individual's history, as well as accessing pre-existing information such as medical or school records. Evaluating case studies In some ways, the findings from case studies are highly valid, as the individual is explored in great depth and within a genuine context such as their work or family ‘Validity may be improved further usin, where the us of diferent techniques should profece similar findings, for example observations and interviews with the participant and questionnaires for their family should all lead to similar conclusions. KEY WORDS ‘<€a8@28002Y: a research method in which a single instance, e.g. one person, family or ing is studied in detail eRHISMGWUISHIOR: is when different of observations, interviews and te |_ study the samesphienamenoR. ff similar results, this suggests the ——— japant’s behsiout ng out about the part siepanicpant’s gamle. Sn ° ia al punion, tbe socal ae meen NN ast as welt “Theat thinking and these ‘emotions as well relatives to fill out “pquestionnaire. If the sane results Mera geal RON SETAE nent oat ge methods SHEE havo se heaccopment otc Sad 2 all te conederaton 9 PARE bem reper This may make te seus af ane one Parga Ne ip wth the Fr outook, whieh woul ce ments hal BANE sabi et {nformed consent, and particular attention should be Muimvaded and that siggy, The level of detail we the questions asked nie pracy 18 nea ncxpants pave fe and TREY Me fase to answer them. Th paid to ensunin Confidentiality o being able 1 deta) In addition evalyate the case tad study This me) et nae Sitanstateriedarts, ment Miter enya NY BO Rinse which Could involve Bing dtl PS the participant (OF sould isk breakin tegmideline oF confidential instance such as & family of insutution) YOU would instance ig nants {echmigues you wou! ecg ga, You would als need 10 5 toexplain wang ould plan the analysand ater ANS of biny ao an sug sateen arc etal peo onl few researchers TNS 0 sod eth et with the ence, reas ha YT the yet aj that i 0 take an EXE find ard of the nang for example when IY unbiaet at heparpant has ui: So irene sath a rly external viewpoint might interpret the findings differently. Tn addition, the causal Piguceed mtr peri tothe individual beng staied 50 Mewes ‘Apsychologst ina sleep cineca bo wg study on a patent, PMA. Saige wy rove few ter instances orfonensat —Coneeet} ot 2Pemerabsabliy of findings from acasestuly whe He ng cary baa dreams for several Yea isvery simi win has had veep labratry where 689 Toeents can be vdeoed to record shaven i pater Seeptalng and sleepwalking, The Applying your knowledge of paychologist uses an BEG elecvoencepralcare” : Payenet a that records brain waves) 10 flow case studies to novel research Fame sleap cycies and to detect when the patient situati Penne tthe patient is then woken uP 296 tuations + ae bocttherr emotional state and what he ‘You should be able to recognise case studies, and when anes about A tests also conducted when itis appropriate to use one. You should also beable to qeerraent snot drearing, they are again wiser aaasePzibl techniques that could be used incase tio Sd asked about their emotional tate FYI Pe aren asked to Keep 2 dream dary tO record Study. When making these decisions, its important toconsider the validity and reliability of the findings ‘One way that the validity can be improved is through viangulation, where different methods are used within problems wath nightmares started and how anes the ease study to obtain the same information: for they occur ra eathe bad dreams occut. Members of PMS “erat have been interviewed to find out when he CAMBRIDGE INTERNATIONAL AS & A Leve { you can, work in par Individually, look atthe ice box and answer | search methods n pract | the following questions or share the question | berween you with one person doing the pumbered questions and one person doing the ‘odd-numbered questions 1 How many different methods and techniques can you identify? 2 Give two reasons why this is a case study 3 The psychologist is concerned that if the patient PMA wanted to withdraw from the Prods, this might be quite difficult. Explain why this might be so. Explain the ethical reason forthe researcher 4 reterring to the patient as PMA, 5 Suggest one other ethical issue that might be 2 problem in this study. 15 Suggest one methodological issue that might ibe 2 problem in this study. Peer assessment ook at the answers given by our pais, inven e sider the following point for your partner. Con each question: 1 each of the ideas given a research method Bs technique? Are there any others? re are only two features that make a 2 eet case study. Do they have them both? ‘atures, which are not strictly study, also included? 1 ‘Are other fe processary for 8 caSe isthe explanation complete? Check the reason given. It should be 4 eretidentilty not ‘privacy’. Why would ‘privacy’ be incorrect? ethical issue suggested relevant to this 5s, why? not, why not? methodological issue suggested the : eet to this study? If so, why? If not, why not? PSYCHOLOGY: COURSEBOOK Questions 6 Damon and Inka are planning a case study of 2 person living with a Usability. t0 investigate responses 10 situations imsohsing help. a Suggest three techniques that Damon and Ir might use in their case study bi Describe two ethical problems that mig] ane inthe study that Damon and Inka are planning ii For one of these problems, suggest a Possible solution. ¢ Explain whether the results from Damon and Inka’s study would be typical of the way everyone would respond to such situations 1.4 Observations (Obsseryations involve watchin ! ‘There are many ET ce: that observations are conducted and we will consider eat intam, Anguatmeanal tes choices is important both for understanding and planning obgervationalstadies Thosettingiia Which the observation takes place is one such choice: A-gaturalistiSGESERfinn is conducted in th panicipants’ normal environment, wilSUFIMEATerenCe From te eset incite th wiakorpiaal eaviFORMERL. A controliedmatserratjon is conducted ina situation that aS becn manipulated by the researchers ‘This may be in terms of the social environment, such as varying gr0Up Sizes or adding a model, or in terms of the physical environment, such as providing objects for play different fo0ds or new locations. Controlled observations ‘can be done in either the participants’ normal environme ‘or in an aruficialisituation such as a laboratory 1: a study conducted by watching the participants’ behaviour in their without amar the researchers in either the social or physical ‘environment coampalagpabemnnnten: ‘a study conducted by snatching the partcipants’ behaviour ina situation in which the social or physical environment has been manipulated by the researchers. It can be conducted in either the panicipants’ normal environment or in an artifical stuation hh unstructured "@b¥€PVEHOR: a study in which the le range of possible usually confined to a p ng of a study to refine the jories to be observed structured-observation®a study in which inited'range | the observe behaviouralcategories: the activities Yed in an observation. They should be 1 (clearly defined) and should cecal 2 continuous streammof Bctiity intoiserate fecordablerevents. They must be observable Scions rather than infemadistates [tthe beginning ofa study, observations may be non- focused, that is, the whole range of possible behaviours is considered. If this continues throughout the study, itis called arvunstructured observation, More often, the range being studied is narrowed to a fixed set of behaviours, a technique is called a structured observation. The specific activities to be recorded are clearly defied in behavioural Categories. This helps the observers to be consistent, ic it improves ihtersobserver reliability. Another decision to be made is the role of the observ in the social setting. This may be participant or non- participant. A is part of the social operationalise Setting so may, for example, engage in conversation with adults or play with children. Aw does not become involved in the situation being studied. This can be achieved by watching through one-way lass or by keeping apart from the social group of the participants, for example by sitting in a separate place such as a bench in a park. the consistency between two researchers watching the same event, i. whether they will produce the same records researcher who watches Fom the perspective of being part of the | social setting Per Eescome involve aresearcher who does | the situation being studied | 1 ch methods The r is obvious hidden or disguised so the participants do not know individual is an observer). A participant observet would be overt, if they were holding a clipboard. When + is disguised as a member of the ‘a non-participant CTY), a participant obser social group (Figure 1.9) or whe observer is physically hidden (eg. by using they are covert | Participants cannot be aware that they are being watched if the observer is covert. This increases validity roi is unlikely that participants would be affected by being observed, so demand characteristics and the effects of social desirability are reduced. However, ise practical issues, as the observer covert observations ra thust be either hidden, far away or disguised in their role This may make data collection more difficult, potentially ind reliability. Furthermore, covert reducing validity a as the participants observation raises ethical issues cannot give informed consent, and if they work out mnt observer's role this can cause distress a partiipal Stseret: the role of the observers fo the participants ‘eovertiobserver: the role of the observer is ot obvious, e.g. because they are hidden eaguced Figure 1.9: Acovert participant observer is disguised by being part ofthe social group > Evaluating observations Naturalistic o turalistic observations have the advantage that the behaviours seen are true to life. They are more reflect the way the individuals really behave then fhe is interference inthe situation from researchers as isthe case in controlled observations. However, there is no suarantee thatthe behaviours bein studied will actualy ‘cur ina naturalistic situation, so it may be necessary to use a controlled observation, : Using an unstructured observation ensures that any important behaviours that happen are apcommtsed and recorded. However, it may be very difficult to collect data on all the activities agouratelwand many behaviours may betamlamaide the aims of the study. Itis therefore likely especially when only epesifiometivitis are of interest, that a structured observation will produce more reliable data as the observer will only need to focus on a small number of categories. The role played by the observer, and the participants’ awareness of this, affect validity If participants are ‘unaware of the observer, or can ignore them, theit activities are more likely to reflect their normal behaviour. This means that covert observers would Produce more valid results than overt ones. However, in the case of covert participant observers there is an ethical issue of deception. The participants may interact ‘with the observer in ways that they might have chosen not to if they had been aware thatthe individual was a researcher. This could invade their privacy and cause harm by distressing them. However, an overt observer ‘may alter the behaviour of the participants as they are aware that they are being watched. This would reduce the validity of the findings as the activities being. recorded are less likely to reflect real-world behaviour. Applying your knowledge of observations to novel research situations (Observations can be used either as a research method of as a means to collect data in other research methods such as case studies, experiments or correlations. ‘You should be able to distinguish between these two situations. Observation is being used as a research method in itself when the study consists solely of a means to collect data by watching participants and recording their behaviour directly to provide data. Observations are used as a technique to collect data CAM v1 SRIDGE INTERNATIONAL AS & A LEVEL PSYCHOLOGY: COURSEBOOK about variables in other research methods when they are used to measure the dependent variable in an experiment or one or both variables in a correlation. In a case study, observations can be used alongside other techniques to. explore a single instance in detail, ‘You will need to be able to decide when itis appropriate to use observations as a method, or as a technique within other methods You should be able to chores about to be able to suggest ways of achieving these, such as how ‘to make an observer participant, or covert In addition to being able to select and evaluate the ‘observational method and the different techniques, you should beable to plan studies using any of the techniques in.a way that is appropriate to a given aim. All ofthe ‘technique’ decisions (eg. naturalistic, structured, participant, overt) need to be made for every observation, So, you may need to choose and describe the location for a naturalistic observation or how the environment will be controled; to be recorded ina structured observation; how observers would aim to be included inthe socal seting (participant observation) ‘or avoid the socal setting (non-partiipant) and how an observer could be made covert Se easing ae Dr Huang is interested in whether her students detect each others emotions and plans a controlled observation. She asks tree students to act 2s confederates. They are told to take iin tums to appear quite sad in the common room at lunchtime | Dr Huang ison lunch duty wth a colleague so they ‘ean act as non-partcipant, overt observers as they walk through the common room. The students will |, take no notice of them as they are used to them being there. Dr Huang suggests that she and her | colleague use a lis of specific behaviours to record, 0 they are doing a structured observation. This wll also help to raise interobserver raiabilty as they will be working from the same definitions, such as recording shows concern without action’ if people look atthe confederate without moving towards them, if someone goes up to the confederate and asks them ifthey are OK and ‘takes action’ ifthey engage in a behaviour such as putting their arm round the confederate of buying them a drink a a | 2 > Look at the Research methods in practice box «Suggest one other behavioural category that might have been included in the structured ‘observation and define it If Dr Huang had conducted an unstructured ‘observation, how would the method have differed? ‘An alternative plan would have been to have conducted a naturalistic observation and watched to see if any students appeared to be sad and how others responded. Outline one way it would have been ethically more acceptable and one way in which it would have been ethically less acceptable. Suggest how Dr Huang could have used a participant, covert observer rather than being an overt observer. Questions 7 Debra and Jin want to use observations to find ‘out about the behaviour of animals. Debra wants to go to the park and hide in a tree to observe the animals that lie there. Jin thinks it would be better to set up an artificial situation and watch laboratory rats interacting with objects they would put ina special box ‘a Who is suggesting a naturalistic observation and who is suggesting a controlled observation? b Explain whether the observers in Debra’s study would be overt or covert. ¢ Foreither Debra’s or Jin’s suggestion, decide whether it should be conducted as an unstructured or a structured observation and justify your choice. 1.5 Correlations A correlational analysis isa technique used to investigate a link between two (measured variables). Gonzelatons are useful when itis possible nly to measure variables, rather than manipulate ‘hem, ie. when an experiment cannot be conducted: This may be because changing the variables would Not be practical or would be unethical. For example, it would 1. Research methods not be practical or ethical to conduct an experiment which controlled children's longsternfexposure (0 television and it would not be ethical to increase real- life exposure to violent television programmes. Both of these could, however, be investigated using correlations Real-life exposure of a group of children to violent sie ‘could be measured and correlated with another measured variable, such as how aggressive their behaviour was in school. = “-cowariables: the two measured variables in ‘a correlation ‘eorretation: a research method that looks for a relationship between two measured variables ‘A change in one variable is related to a change in the other (although these changes cannot be assumed to be causal) ‘causabirelationshipea link between two variables such that a change in one variable is responsible for (i.e. causes) the change in the other variable, such as in an experiment ‘To look fora correlation between two variables each variable must exist over a range, sometimes called continuous data',and it must be possible to measure them numerically. Suitable variables would include durations of time, totals of tallies aumbers from a rating scale or test scores, Several techniques can be {ised to collect data for correlations, such as self-reports, bservations and different kinds of tests and tasks Itisimportant to recognise that any link found between two variables in a correlation cannot be assumed to be causal relationship, that is, we cannot know whether the change in one variable is responsible for the change {nthe other variable (Figure 1.10). Ths is possible in the vise of an experiment, as we can know that there is @ ‘ bt & st EvEnuEe € ¥ COUPSEBOOK A Questions to embark on a course of treatment. such as a new drug 8 EAus and Tok ae regime or people experiencing a hfe change, such as at has and Talos are goung to ind mut f ther 3 IBeepinnng GPR pregnaeey. Sortelation beeen te amount of cffe people ber of dreams they recall 3 Explain why th not an expenment tional study and Suggest whether the results will show a positive correlauion or a negative correlation. © i Su IgEest OMe Way to measure the amount of cotlee that people drink. Explain either one strength or one weakness of the way you have suggested ‘measuring this vanable. 1.6 Longitudinal studies A is one that follows a single group of participants over time, studying one or more variables - a By measuring variables over weeks, months, years or even decades, a longitudinal study can detect changes in individuals. People change over time, not just as they age but as they accumulate life experiences. In addition, longitudinal studies can explore the effect? of specific ‘experiences’ such asiteryentios or events! ondevelopmenty The use of longitudinal studies is an alternative to ‘cross-sectional studies as a way to look at effects over time. In differences betweell people of differing ages, or at diferent stages (such as! different periods after an intervention) are explored by* comparing different groups at one point in timé For ‘example, to explore changes in WHiSQMUIRERS with age, 2 cross-sectional study could look at 10-,20-, 30-, 40- and $0-year-olds (Although a cross-sectional study allows researchers t0 investigate dongstermaniiuences, it is difficult to separate changes over time from differences due to the individuals growing up at different times. For example, if a cross-sectional study looking at helpfulness in people of different ages found a pattern in the results, this could be due to differences in upbringing at the time ‘when they were children or the changing expectations of Society rather than their agefSuch problems are avoided « jna longitudinal [Link] a a groupof participants sare articipants all selectp aestuneme Mpeorsta, chosen and tested at different time pomtg A cohort + ‘might be individuals born in a certain year, people about Loupdina sendin may hae ications or deagne Some es COCO TSEPOPE in sarates wo inves atetbem, ch a BATE tanouons or belts as we grt olde, Other stds record longitudinal changes in two or more vanables to investigate possible comets Between thera, such tSenponee i2a aultcaaral ciety ood prejudice. Yetotber we an experimental design whee one or of time, are expected to change as a consequence of a deliberate manipulation. In the last case, the Vanables being measured are dependent variables and the testing at different time points in relation to the deliberate manipulation is an independent variable. To provide a baseline for tracking long-term change, the variable is usually measured at the outset of the study, and again after one or more periods of ume. This procedure can be thought of as a very long-term repeatgd measures design, so is sometimes referred to as “gesSimamenl This procedure can be usedcTor example, to measure the consequences of interventions such as educational healthy mental iealth{ ‘or occupational ones that aim to improve outcomes ‘One level of the independent variable is the baseline test or’ time point and one or more repeat testing sessions occur, providing a ‘posteimteryeation’ level of the independent vanable and possibly more time points during or long after the interveation as a follow-up. (USRGRUSEPHURY: 2 research method that follows a group of participants over time, weeks to decades, looking at changes in variables to explore development or changes due to | experiences, such agintenventions, [Link] therapies compares people at | different ages or stages by comparing different | groups of participants at one point in time egohomaa group of participants selected at the J | ee Evaluating longitudinal studies ‘A longitudinal study has the clear advantage over a fross-seetional study in that researches can be confident any changes they detect are due to the passage of time and not the effect of differences between the cohorts of participants representing each age or stage. This has two benefits Loma diem, ie repeated measures ‘designs retest the same individuals. This means that the researchers can be confident that differences are not {ue to participant variables A second consequence is tat it also overcomes For example, in the Terman study of highly intelligent ‘children (see Figure 1.11), differences in educational pathos cre oppor unis and other octal con {that could have lematic in a cross-sectional tidy would not influence this longitudinal study. ‘Confounding variables such as this, which are aspects Of the environment, are called situational sariables. This imethod is therefore the most valid test of developmental ‘changes Figure 1.11: Lewis Terman, who began a long-running longitudinal study in 1922 1n experimental design Where the same participants are tested on two or more occasions over a long time, e.g. before and after a six-month intervention le: a confounding variable caused by an aspect of the environment, for example the amount of light or noise ‘ataligeneequotiantiiGlng measure of general feasoning and problem solving ability 1 Research methods A very long-running longitudinal study isthe Terman Lie Gye Say of Chien wit ign aan which began in 1922. Although the study eventaally debunked the myth that a high intellgenceqemtient ‘WQvinevitaby leads to success, some of is parti did become famous such ashe prycbologit Lee Cronbach, who specialised in psychological testing and measurement, became president ofthe American PsychologicalAssociation. However, precisely because longitudinal studies lat fora long time, the sample sie ikly to fal this iscalled ‘Terman’ initial sample, in 1922, was 1528 children, Ina compas between dala collected in 1996 and 1999, although 162 questionnaires were returned in 199 just 119 partipants provided data on both occasion (Hoan & Velasquer, 2011), Where did they go? Participants may levestadies for many reasons, but thes tend to be more apparent ina longitudinal study. They inciode- growing older and deciding to withdraw consent (eg when consent was orginally given bya parent, boredom (Com bing repeatedly tested), moving home and becoming uncontactabl life events that prevent participation (oor physical or mental health, lack of time becoming homeless or going to prison) and, ultimately, dying. ‘[Link] loss of participants from 2 sample overtime. This ay be de to many | reasons, such a losing contac, the deseo | discontinue, for example though boredom, being Unavailable or death | ‘A smaller sample is likely to belessrepresentative, but especially so in this case. Sample attrition means that generalisability is progressivelytareatened as the sample ‘becomes concentrated on willing, motivated, people who live stable, healthy, law-abiding lives. These people may not be the most psychologically interesting! The people ‘who stay ina longitudinal study may even do so because they feel ‘special’ by participating, creating an issue for validity. For example, Terman’s participants knew they hhad high 1Qs and the sense of being important was reinforced by having a name: they were referred to as Terman's ‘Termites’ Participation could, for example, have given them a motive to compete or succeed, or & sense of community ie. the very act of participation will had made these individuals different from non- participants, even ones of similarly high intelligence. REESE a) CAMBRIDGE INTERNATIONAL AS & A LEVE Another problem caused by theghurationot longitudinal studies relates to re! lit. To ensure thal data collection at each time point is consist the measures used need to be standardised, Ho over tume, ways to measure variables may be improved. Potentially leading to the need to introduce different better measures. In addition, researc the study may move on so, f hers working on for example, the individuals administering interviews may not remain constant, Conversely, if researchers continue working on a study for a long period of time, they are likely to get to know the participants if they have direct contact with them The changing development of these rela nt of these relationships can also be confounding variables i Finally, longitudinal studies can present particular cthical issues For studies starting with children, obtaining consent in a way that the children understand ’s desirable, as well as gaining consent from their Parents or guardians. Furthermore it is harder to decide whether a child wants to withdraw from a study than itis with adults, who are more likely to speak up for themselves. Participants must also be repeatedly asked for consent (and reminded of their right (o withdraw) and must be recontacted at each time , point. This requires the maintenance of an extensive bank of contact details, which presents an issue for confidentiality Applying your knowledge of longitudinal studies to novel research situations Longitudinal studies provide information about long-term changes. They are therefore suitable for that relate to development and ageing, but also to any other life changes. Such changes include both those that are the result ofa life event, such as recovery from an illness or the effects of a marriage or divorce, and those that are the consequence of a manipulation, such as attending a course or planned. ‘exposure to social groups. You should be able to identify the variables being measured and whether these are simply being tracked aver regular time points to monitor ible changes. or whether there 1s a manipulation nfluence Which 1s acuively intended to cau In the latter ca Id be "before, durin after and follow-up’ in relation to the manipulation This situation Would be a Igngirudiastudy wi an experimental design, time points ¢ You will need to be able to decide when to use a Jongitudinal study rather than any other method. and which type of longitudinal study is appropri in a given situation. You should also be able suggest variables to study and ways to using a test, questionnaire, interview of observation, and how the results could be analysed and used In addition to being able to select and evaluate the longitudinal method, you should be able to plan a longitudinal study in a way that is appropriate to a given aim, such as whether or not there is a manipulated independent variable. Remember that you need to identify one or more variables to investigate, consider hhow they will be measured, and decide om the intervals between the testing time points and how you wil standardise procedures to ensure consistency overtime as well as planning how to re-contact participants to retest them Questions 9 Hanif is planning a lo changes in professional ability as adults pr through their careers. He has a participants from different types of employment and has various measures of cai such as how much they earn, the senionty of their position in their organisation and how many stalT in the organisation are above and below udinal study to measure a Suggest how often Hanif should cont. participants to track their progress. b Hanif wants to measure how happy they are in their career. Suggest how Hanif could measure this. ¢ The participants will need to be recontacted cach time Hanif wants to test them. Si how Hanif could do this 1,7 The definition, manipulation, measurement and control of variables variables are factors that change or can be changed. In Mettriments these are te independent and dependent abibles as wel as variable that ae or are not ontroled, In correlations, there are two measured ‘atiables called co-variables _ Baperiments look for changes or differences in the Mependent variable (DV) between two or more levels ye: independent variable (IV), which are set up «By the experimenter. tis important that the IV is © gperationalised, so that the manipulation of the Ronditions represents the intended differences. At is sere of tis chapter, we discussed the idea BE the afm of a study, that is, what the researcher is ying to find out. The aim refers to the variables but ot necessarily very clearly. Operational definitions of ables help to add clarity to the intention expressed the aim, Consider a study testing the effect of age on usceptiblity to false memories. The TV would be age, for example, ‘young, ‘middle-aged’ and ‘old’ {BrOUp It is important to know how old the people in the groups are; this is operationalisation. You might ‘perationalise ‘young’ as under 20 years old, ‘middle- ‘aged’ as 40-50 years old and ‘old’ as over 70. The DV Bist also be operationalised, soit can be measured effectively. We could operationalise the DV by counting the number of details ‘remembered’ about the false fiemory or how convinced the participants were that it was true, KEY WORD aim: the intention of the study, the idea being tested or the such as to investigate a question or solve a problem 1 Research methods Dr Huang is thinking about buying new chairs for her classroom. Het aims to explore whether hard Gr soft chars help her students to work better (Figure 1.12), She wonders whether to predict non-directional (two-tailed) hypothesis: There is 2 difference in work rate of students sting on ‘comfortable and uncomfortable chairs. Another psychology teacher says that students respond well to other comforts ike access toa drinks machine or snack bar, and the soft chairs might ‘make them happier, so they work harder. Or Huang rewrites her prediction as a directional (one-tailed) hypothesis, saying: Students on comfortable chairs wall have a higher work rate than ones sitting ‘on uncomfortable chairs. third teacher is not Convinced and suggest that if the students are too comfortable, they wil become sleepy and lazy, 0 «work ess. The hypothesis would then be: Students «on comfortable chairs will have a lower work rate than ones siting on uncomfortable chairs. Ths s also a directional (one-tailed) hypothesis. Her null hypothesis in all cases) would be: Any difference in work rate of students sitting on comfortable and Uncomfortable chairs is due to chance. ‘Now imagine a study that aims to look for a link between sleep and emotions. A non-directional hypothesis might be: There will be a correlation between amount of sleep and emotional reactivity. A possible directional hypothesis could say: There will be a positive correlation between amount of sleep and how emotional someone is. Note that the hypothesis does not say that one factor causes the other to change. The null ‘ypothesis here would be: Any correlation between ‘amount of sleep and emotional react ty is due to chance. Figure 1.12: Would you work harder in lessons if you had more comfortable classroom chairs? Controlling of variables Psychologists need to control variables in their stud tebe more certain aout he ince oe salidty In particular, in experiments itis impornnt to that i, ones that have 8 consistent effect 0 they confound, Le ceo ne results, Confounding variables act on the DV selectively in one level of the TV so can interfere with the recite i one of two ways. One possiblity i that they ean work against the effect of the IV, counteracting its effect on the DY, so preventing areal effect from being identified Alternatively, they can increase the apparent elfect of the TV and so suggest that there isan effet that doesnt really exist. These variables are the most important to control Other uncontrolled variables, which have a random effect on the DV across all levels of the IV, are not so Problematic. The difficulty isto identify which variables Will be important to control before the experiment starts. This is one function of a pilot study, a preliminary test of the procedures of a study. The intention isto identify potentially and control them. This prevents or limits their effect to avoid ‘making the results difficult to interpret if it is hard to separate their effects from those of the TV. Consider Dr Huang'’s study of students and chairs. Perhaps Dr Huang compares one class in a room with the new (soft) chairs and another class in a different room with the old (hard) chairs. If the room containing the new chairs happens to have better lighting, Dr Huang may find that the students in the ‘soft chairs’ condition perform better. However, this may be due \o the confounding variable of brighter lighting rather than the comfort level of the chairs. This is an example of a situational variable, because lighting is an aspect, of the environment. Another possible confounding variable is how hard-working the individual students are. We might expect normally hard-working students to be randomly distributed among the different classes, in which case this variable isnot a problem. However, suppose that all the students in the ‘soft chairs’ class do arts and humanities subjects and all the students in the ard chairs’ class do maths and sciences. If Dr Huang A LEVEL PSYCHOLOGY: COURSEBOOK happens to use a test of data analysis as her, the DV, she might find that the students in the “hard. chairs’ level of the IV perform better. This would suggest that the soft chairs make students perform worse but could in fact be due to the uncontrolled variable of subject groups. This is an example of a participant variable, because the difference has been caused by a feature of the individuals, ie. their ability in maths Standardisation Controls make sure that the levels of the IV represent what they are supposed to, ie. that the differences between them are going to create the intended situations to test the hypothesis. This helps to ensure validity. tis also important that every participant is treated in the same way. This is the process of standardisation. One way that this is achieved is by having standandined instructions that give the same advice to every person in the stu). Imagine a questionnaire testing attitudes to helping behaviour. All participants would need to have the same advice about how to fill it in, so that any effects of social desirability-the influence of needing to give answers that were acceptable to society-were equally likely. ‘Seafdareisediinstruetion® the written or verbal information given to participants at the beginning and sometimes during a study that ensure the experience of all participants, regardless of level of the IV, is as similar as possible The procedure itself also needs to be standardised. This involves having equipmeat or tests that are consistent, ice that measure the same variable everytime and always do so in the same way. Consider the questionnaire about attitudes to helping again. Al the questions should focus on the same aspect of behaviour, ie. helping, rather than some looking at a different but possibly related factor, such as being friendly or happy. In Taboratory experiments, standardisation is easier than in other studies, as equipment is likely to be consistent, for example stopwatches or brain scans. However, some of these measures, such as brain scans, may need to be interpreted and this must also be done in a standardised way (Figure 1.13). 1 Research methods “igure 1.13: Scientific instruments are likely to produce ive, reliable data ditions (to achieve validity) and to measure the DV consistent (reliable) way, and to be able to justify ee choices You will also need to beable to decide hat controls it is appropriate to use and to suggest how ese can be implemented. a Pilh Or Huang’s experiment about classroom irs, the IV of hard and soft chairs must be Operationalised. The text also referred to them 8 ‘comfortable’ and ‘uncomfortable’ chairs, but fthis still does not make clear what is meant by figte'c! and ‘soft’. This could be done by saying Pe chairs with wooden/plastic seats’ and ‘chairs with Padded seats’. Operationalisation of the DVis also Needed. The text referred to working ‘better’ and ‘harder’ but this is also incomplete. We need to ‘expand on the idea of work rate, which was also. Used. This might be measured by counting the umber of pieces of homework handed in late, or the time spent doing extra work. Either of these Would indicate the amount of work being done. There are many confounding variables that could be important in this study, for example, some of the students might work harder anyway or the rate Cf work might vary with the weather. f students worked harder on sunny days, this would be a situational variable. The important variables to control are those that could confound the results For example, if there was a choice of chairs, the students who chose to sit on comfy ones might be the laziest. fleft as an uncontrolled variable, this could alter the results by making it look as if soft chairs made students work less. In the description of the correlation on sleep and ‘emotions, the co-variables were the ‘amount of sleep’ and ‘how emotional someone is’ or their ‘emotional reactivity. It is important to operationalise variables in correlations too. An operational definition of ‘amount of sleep’ could be the hours and minutes spent unreactive, or could be identified with an EEG indicating rapid eye movement (REM) and non-rapid eye movement (nREM) sleep. A questionnaire could be used to measure their emotions, with an operational definition of ‘emotional reactivity’ as how likely @ person is to respond negatively to innocent comments, or how often they argue. a Nane Look at the Research methods in practice box. For the experiment: Suggest one other way the IV could be operationalised. ‘Suggest one other way the DV could be operationalised. Would a possible difference between how lazy students were be a situational variable or a participant variable? Suggest one other possible uncontrolled variable. the correlation: Suggest one other way the variable of ‘emotions’ could be operationalsed.

You might also like