comments for 18 th of april
http://kpsychb.wordpress.com/2011/12/09/the-importance-of-obejectivity/
http://ryan1392.wordpress.com/2012/02/12/ethics-needed-or-a-road-block-in-psychology/
Archive for the ‘Uncategorized’ Category
operationalizing variables
Variables are really important in a psychological research as they do specify our results. How we manipulate our variables (such as behaviour, characteristics) will have a result on the outcome of the research. In order to gather the desired outcome we need to operationalise those variables. This means that we need to define in depth our independent variables and dependent variables. In other words we have to say how we want to measure the behaviour. For example Bandura (1961) intended to measure aggression. We cannot simply just measure aggression as there is different types of it. Therefore he specified the inter rater observer validity: such as kicking, punching the doll where the behaviour perceived as aggressive. There were three conditions in which children could either see a model who was behaving aggressively do the doll or not. The children were them observed if they would repeat aggressive behaviour or not to the doll.
Some of the behaviours cannot be seen and therefore we have to operationalise them in order to investigate the reason underneath. For example thoughts in depression cannot be overtly observed. Therefore we need to give for example a valid questionnaire which measure thoughts (independent variable) to see what thoughts people with depression have it (dependent variable). If the research operationalise variable in a right way it allows him/her to exclude any extraneous variables/confounding variables. This means that all the unwanted variables that could affect the results are controlled and do not affect the explanation of the behaviour. For example if our independent variable is a memory task, other variables such as: participant’s tiredness, mood etc could have an impact on the results (dependent variable). Therefore it is extremely important to strictly operationalise our independent variable. Good control over the variables is also beneficial to the research as it shows that the research is high in internal validity.
On the other hand operationalisation of the variables could be a bit subjective in some methods. Take into account natural experiment or observational studies. Yes you define the inter rater reliability (you operationalise what you will measure by specifying the patterns of behaviour that you look for) however every researcher have got they own scale and reasoning on what they perceive as right. This could lead to the researcher bias. For example, Rosehan (1973) conducted a research about people behaviours after labelling mentally ill people with different disorders. He and his colleagues were admitted to the psychiatric hospital on the basics of schizophrenia disorder. In fact they were perfectly fine they were just faking the symptoms of that disorder. The aim of this study was to observe how the medical staff treats people based on the label imposed. Obviously the behaviour that they reported was only observed by them and therefore the independent variables was not really operationalized. Clearly this threats internal validity of any research.
In conclusion the purpose of every research is to operationalise it variables in order to minimanise any errors that could possibly affect the findings and give wrong results. Researchers strive to make their research looks high in internal validly(good control) by sacrificing external validity (generalisability) . Every research design have main two important variables that are being defined (operationalsied) . Independent variable is being manipulated to see the effect on the dependent variable. Different factors investigated will have different operationalisation.
Comments week 8 :)
1.
http://cerijayne.wordpress.com/2012/02/18/should-children-be-used-in-psychological-research/
2.
http://elrucron.wordpress.com/2012/03/04/manipulation-increasing-false-findings/#comments
3.
http://raw2392.wordpress.com/2012/03/02/should-animals-be-used-in-research-110312/
4.
http://lrowlands1.wordpress.com/2012/02/19/accidental-discoveries/
🙂
Discuss the use of laboratory experiment in accordance to assumption of cognitive approach.
Discuss the use of laboratory experiment in accordance to assumption of cognitive approach.
Mind Uploading – this short metaphor would probably the best describe the method used and assumptions of the cognitive approach. Cognitive approach assumes that our behaviour is mainly based on how our cognitions operate. Cognitive psychology uses variety of different methods to investigate its assumptions, such as case studies (the famous case of KF). Although the preferred method in this discipline is laboratory experiment as it provides most scientific evidence supporting the assumptions. In other terms is a computer science discipline, that’s why metaphor used at the beginning is completely adequate. Let me explain why.
On the contrary to other psychological approaches such as behaviourism (that is based on observable behaviour), cognitive psychology focus on internal mental states.
‘’all processes by which the sensory input is transformed, reduced, elaborated, stored, recovered, and used. It is concerned with these processes even when they operate in the absence of relevant stimulation, as in images and hallucinations… Given such a sweeping definition, it is apparent that cognition is involved in everything a human being might possibly do; that every psychological phenomenon is a cognitive phenomenon.” By Neisser (1967)
One of the cognitive approach applications that laboratory design has been used for, its on memory research .
“Memory is the process of maintaining information over time.” (Matlin, 2005)
As memory is an internal state it cannot be measured by any other method than laboratory experiment. This is simply because we cannot observe memory but we can test for example :the recall of it in accordance to the theory of how we process information.
But what it is laboratory experiment? It is a procedure carried out in laboratory setting where all the variables are controlled. Therefore it is highly scientific as it can be replicated and extraneous variables are less likely to provide other explanation of the behaviour investigated as they are controlled as well.
Atkinson and Shiffrin, (1968) are well known in psychology for proposing the simplified model of memory (multi store model). They assumed that our memory poses couple of stores. How we process the information depends how the information flow through those systems. Firstly information enter sensory memory (information are stored for couple of seconds) them the information are passed to the short term memory (stored for 18 sec and we encode 7 information) them the information are transferred to the long term memory through attention and rehearsal. Long term memory has unlimited capacity for both storage and remembering information.
Murdoch (1962) investigated that theory using laboratory experiment. He presented participants with list of the words and asked them later to recall how many they could remember. He found that the words presented at the begging were better recalled (primacy effect) because they were transferred to long term memory. He also found that words at the end were better recalled (recency effect) because they were in short term memory.
There are lots of advantages to this method (already mentioned before). Such as scientific validity, good internal validity etc.
On the other hand this method has been criticised a lot for being low ecologically valid. This is because the artificial setting is created in order to investigate its theories. This could be off putting but we have to bear in mind that this method its not the only method that cognitive psychology uses. For example field experiment (eyewitness theories by Loftus and Palmer 1974) . Therefore it can be argued that cognitive psychology as a discipline is high in ecological validity.
Conclusively despite the disadvantages of the laboratory experiments as a method, cognitive psychologists still uses this method as main when investigating assumptions of the cognitive psychology. It is the best method to investigate the internal states because they can be objectively measured by using laboratory experiments.
Is it ethical to collect the data from internet to carry out the psychological research? Week 4
Is it ethical to collect the data from internet to carry out the psychological research?
In my opinion it is perfectly ethical to do so. I do not think researcher would violate any serious ethical guidelines. If the research risked and decided to carry out the research through internet them I think he is taking a greater risk of his research failing to establish any cause and effect relationship rather than hurting someone by using personal information. I will explain why by considering 5 ethical guidelines imposed by APA.
- 5 ethical principles – informed consent
- Protection from psychological harm
- Right to withdraw
- Confidentiality
- deception
As some of the people would straight away agree that is strictly unethical and distress provoking to carry out research online, by gathering data from forums, chat rooms and many others. I would still argue it is not. First of all, people who are willing to write on the forums comprehend that they are writing at the public domain. By public domain I mean that anyone from human population is allowed to access the website and read it. They have to consent to terms and conditions of the website (before posting anything) therefore any member of the public not only researcher can access and use the data presented there, in any way. Therefore the criterion of informed consent is satisfied. People have choice to make their post and blogs to be private and if they do not wish for their data to be available only for them, they automatically consent for their data to be used. I know they do not directly consent for their private data to be part of the psychological study but all data is anonymous anyway. Some of the quantitative studies did not satisfy this criterion and they were still considered ethical. For example Sherif (1972), subway study. The confederate of the experiment dropped on subway in New York. They tried to investigate the bystander effect. Obviously they could not gather any consent from participants as it would confound the study. The study could cause potential psychological harm in participants due to feelings of guilt of not helping etc. Therefore if we gave our online participants informed consent would it not influence our findings? The blogs have got options to review the posts and change. Would the potential participants not want to show themselves in the best light? Furthermore if researcher decides to gather data from chat rooms, would it be really possible to obtain informed consent? (As people go in and out of the chat room all the time). If researcher want to gather big sample of data them would be fair on him to send 7000 informed consent across the internet? ‘’Internet people’’ still have got option, if they want they data to be public they can include disclosure on their blogs saying that they do not wish their data to be used or include copyrights policies. Like for example I did on my photography website http://www.facebook.com/pages/Desire-Photography/258625364164258 . (P.S. you can like the wepage 😉 )
On the other hand, I think informed consent should be obtained from people that have some kind of artistic involvement, such as: art, music, film etc. Researcher should not be able to use their data unless asked for. Those people should be acknowledged as they worked hard to achieve what they achieved and psychological research findings could in some kind of way influence their career or well-being. For example: if researcher aims to analyse the paintings posted online and the results shows that the artist have got schizophrenia them it is impossible to obtain anonymity as the particular painting analysed, can only belong to that specific artist who painted it. Therefore the career of that artists could be under question mark.
Let’s consider psychological harm. Many people would agree that people would suffer from distress if they knew they data have been used for psychological purposes. For example: if you were using data that is really sensitive, such as someone’s story of rape. I still do not think using someone’s data could cause any serious psychological harm. First of all most of the people online do not sign by their name and surname as they want to be anonymous, as they could be ashamed of their experiences. Secondly researcher must make sure that all data gathered is confidential anyway and any personal details obtained will be not disposed to anyone whose is not entitled to view it. Thirdly if the people through internet are anonymous, how they can know they data was gathered (people have similar experiences) . As they do not know this, they would not suffer from any harm whatsoever. Furthermore some of the real life researchers caused real harm to the participants and yet when debriefing, participants argued that they were happy to participate in the study. For example Milligram (1964) study into obedience ,where participants were deceived about real aim of the study and many of them suffered from noticeable distress from administering electric shocks to the confederate of the experiment( they did not know at first that this person was an confederate of the experiment). More than 50 percent of them argued that they were happy to participate. Therefore does online study can really cause any harm?
Yes I can agree that ‘right to withdraw’ is not satisfied however I do not think, it is important. Lots of psychological researcher did not use this right and still, were considered as ethical. For example study carried out by Slater et al, (1998), have found that babies tend to gaze more at more attractive people. Did the babies really have choice to withdraw from the experiment? Another study conducted by sheriff and king (1978) did not give that right to animals. They imposed electrical shocks to the puppies. Therefore the animals do not have right to withdraw and lots of research are carried out on them. Does it really matter them if we gather anonymous personal data from internet that will not cause any harm to participants and do not give them right to withdraw?
On the other hand ‘internet people’’ still have got right to withdraw maybe not from the study but from the internet. They can simply delete their blogs or any form of their personal information. If they do so , only research will disadvantage from it. As I agreed in paragraph one and two, if we do not need informed consent and it does not cause any harm them I do not really think so the withdraw from experiment is important.
Confidentiality criterion is already maintained by internet and researcher will ensure that the findings are kept confidential.
In conclusion I think is perfectly ethical to carry out the research over the internet as it does not cause any harm, people consent to term and condition of the server for their data to be public and used. Therefore the data can be used by anyone and if the researcher wishes to use the data them he/she should be allowed to do so.
Analyse the reliability and/or validity of a published report.
Analyse the reliability and/or validity of a published report.
I will analyse internal and external validity of the published report. For this purpose I have chosen Shotland and Straw study. You can find whole description of the study at http://psycnet.apa.org/journals/psp/34/5/990/
Let quick recap on what internal and external validity is. Internal validity is the extent to which experiment measure what it intended to measure. Whereas external validity, is whenever the experiment findings and results can be generalized beyond experiment setting to the real life situations, people, times and places.
Schotland and Straw (1976) conducted an experiment in which two confederates of the experiment women and men were involved in argument .In one of the situation women shouted ‘’ get away from me I do not know you ‘. The percentage of the participants helping was much higher than in other situation where women shouted ‘’ get away from me I do not even know, why I married you ‘’ where only 19 percentage of bystanders helped.
In this study 3 experiments have been carried out to explore the bystander effect. I will analyse first one. First experiment investigated the bystander effect. The subjects were randomly assigned to two conditions. First condition named ‘’Stanger ‘’ required confederates of the experiments that had acting training to play a psychodrama in front of participant. Second group of actors were assigned to the second condition called ‘’married’’. Actors were composed of one male and one female. The difference between two condition was that in stager condition, the confederate of the experiment shouted standardised phrase ‘’ ’ get away from me I do not know you’’ whereas in married condition the actor shouted ’’ get away from me I do not even know ,why I married you’’.
Read full description of the methods at http://psycnet.apa.org/journals/psp/34/5/990/
The researcher took great amount of precaution to ensure that the research achieved high internal validity. The amount of control used indicates the strength of the internal validity. Standardised procedure used allows the elimination of the potential confounding variables. For example: standardised quotes used by the actors allow greater replication of the study. They have also used inter rate reliability in order to measure the helping behaviour. They made 4 categories to which helping behaviour could be assigned. Therefore this allowed researcher bias to not occur and not influence findings.
Although research seems to be high in the internal validity I had spotted one thing that could act as confounding variable. Let me cite
‘’One couple (Team 1) consisted of a man who was 1.70 m, 63.50 kg (5 feet 7 inches, 140 pounds) with brown hair conventionally cut, paired with a woman who was 1.57 m, 49.90 kg (5 feet 2 inches, 110 pounds) with shoulder-length black hair. The second couple (Team 2) consisted of a man who was 1.85 m, 90.72 kg (6 feet 1 inch, 200 pounds) with light brown hair conventionally cut, paired with a 1.70 m, 54.43 kg (5 feet 7 inches, 120 pounds) woman with long blonde hair. ‘’
Previous researchers have indicated that the rate of the helping behaviour depends on the other factors, such as attractiveness etc (Lazarus &Roberts 2007). In this particular study , they tried to investigate the difference btw two condition .The study have found the significant difference , although in my opinion that difference could be founded not because of the conditions but because of the way how the confederates looked like. For example the male confederate in the ‘’married’’ condition could be more masculine than the male in ‘’stranger’’ condition .Therefore participants could be more scared to engage in any ‘rescue’ behaviour to prevent confederate of the experiment to shout at the women. This has been showed in result section. They have found that some of the participants locked the door of the room that they have been in, or turn off the lights off.
‘’ In addition to intervention behaviour, 30% of women who had witnessed the stranger condition had taken protective precautions by shutting the door to the room, turning the lights out, and in one case locking the door. During the subsequent interview these women stated that they had taken these actions for self-protection. This type of behaviour never occurred in the married condition.’’
Therefore the way how the actors looked like could affected participants decision in whenever help or not help.
External validity has been also maintained. The big sample size used allowed the results to be more representative of real life population. The random allocation of the participants to the groups allowed for greater participants characteristics to be spread equally. Therefore makes the sample more representative and the confounding variables less likely to occur. The experiment was based on the real life situation. In real life situations lots of people experience fear and stress when they see emergency situation (Halpern, 1995). Therefore it can be argued that the experiment did evoke real feelings in the participants and therefore showed their real behaviour.
One criticism from my side is that they only used university students. The study therefore can be only generalized to university students and not beyond. Clearly the study does not explain bystander effect as it lacks other participants from different background. As mentioned in the paper, the similar situations were occurring in university campus before and therefore the students could have been acknowledged with ways to help. Therefore the results could be biased due to the participants’ previous knowledge. Furthermore university students are generally more aware as they do poses the certain information, knowledge level that normal people do not poses; they could more or less be involved in helping. This could be the reason why they did not find significant difference btw male and females in helping behaviour.
Conclusively it is nearly always impossible to design study that is both high in internal and external validly. The researcher must decide what is more important. In my opinion this research study did have great amount of control and the findings can be generalized but only to the students. However in further studies researcher should take those limitations into account and fix them.
For Julie Blog 9
I have got a christmas present for you. This week I ran out of time and did not manage to do blog. So you would have one blog and comments less to check and grade. Merry Christmas 🙂 xxx
comments for week 9 :)
Please could you mark those comment. Thank you 🙂
http://blc25.wordpress.com/2011/10/27/why-is-research-important/#comments
camilia92 