Evaluation of Stan the Man’s Shock Experiment (Stanley Milgram)

I’m sure by now everybody is familiar with the infamous Milgram study (1953). If you are one of those people missing out, then feel free to read the link below:

http://psychology.about.com/od/historyofpsychology/a/milgram.htm

After reading around for a blog topic today, I found some interesting evaluative points about Milgram’s experiment. For instance most people (including myself) have blogged about or referenced Milgram’s study with regards to ethics. Yes, Milgram’s study was unethical, not following several obvious ethical guidelines. However after his experiment in a post experiment survey from which 92% of Milgram’s participants responded, 84% claimed that they were “glad or “very glad” to have participated within Milgram’s research. [i] Additionally various participants reported that they were pleased to have taken part in the study. These follow-up results almost free Milgram from any criticism he received surrounding the ethical state of this study. Personally I still think the study was unethical on numerous levels, however as he caused no harm to the majority of his participants I now feel as though Milgram’s study can be regarded as a fair, harm free experiment.  My point of view here is shadowed in various enactments of Milgram’s experiment.

One of these enactments was shown recently on Channel 4 Television, as part of Derren Brown’s Experiment series. The show, which was called ‘The heist’, featured an exact replica of Milgram’s shock experiment. As this study was allowed to proceed, I feel as though this is good evidence that Milgram’s study was not as unethical as people make it out to be.  Due to the follow up survey I feel that Milgram successfully portrayed that the majority of his participants were not harmed during the study, thus making it an ethical study. If the knowledge we gained from the survey was never found, then it could be questioned whether the BPS would allow such enactments to take place, regardless of whether it was for television or not.

A link to the Channel 4 Show by Derren Brown, which can be watched online is below:

http://www.channel4.com/programmes/derren-brown-the-specials/4od#2922164

One final evaluative point I would like to make is that there are many replications of Milgram’s study such as:

• Rod Dickinson’s exact replication (2002)[ii]
• Jerry M. Burger’s replication in which he used female participants, and tweaked a few protocol to make his study completely ethically sound (2006)[iii]
• BBC’s horizon series, on a programme called ‘How violent are you?’ (2009)[iv]

In these studies the results gained are extremely similar to Milgram’s original results in the 1960’s. This shows that Milgram’s results are very reliable as they can be re-tested and show consistent results. Burger in his replication study found that 70% went over the 150v mark, compared to 80% in Milgram’s study and in the BBC’s replication only 3 of 12 participants opted out before the end of the study.

In conclusion I would say that Milgram’s study still seems unethical today, and maybe an exact replica wouldn’t be given the go-ahead from the BPS. However with small tweaks to the research methods or procedure of the experiment the study can be replicated. Furthermore I feel that due to the majority of Milgram’s participants stating that they were pleased to have taken part in the study it removes any opinions that the study was too potentially stressful for participants.

---

[i] Milgram, S. (1974). Obedience to Authority: An Experimental View. Retrieved from http://books.google.co.uk/books?id=MlpEAAAAMAAJ&q=doi#search_anchor

[ii] http://www.roddickinson.net/pages/milgram/project-synopsis.php

[iii] Jerry M. Burger (2009). Replicating Milgram: Would people still obey today? American Psychologist, 64 (1), 1-11 DOI: 10.1037/a0010932

[iv] http://www.bbc.co.uk/programmes/b00kk4bz

 

Blog Entree Week 3 – Do incentives for participants to take part in research affect results?

After taking part in a few studies this week, this question seemed appropriate to blog about. Compensating or paying participants raises obvious ethical concerns, but as ever with ethics there is not a specific set of rules which we have to follow.

The BPS state that, “The payment of participants must not be used to induce them to risk harm beyond that which they risk without payment in their normal life style.” Therefore it does not state that people cannot pay participants in psychological research. However, would I as a researcher want to pay my participants?

Personally no I think that if I am paying for my participants to take part in my study then I am using a bribe to increase my sample size, additionally they might not be willing to take part in my study if they did not receive the incentive. On top of this I feel that I am removing my participant’s right to withdraw, as they may feel inclined to finish the study due to their payment. So ethically I feel as though paying participants to take part in research is unethical. A Famous medical study by Walter Reid (1900)[i], highlights how extreme some people would go to receive money. Reid wanted to test how people caught yellow fever; he paid all his participants $100, prior to the study and $100, compensation to the participant’s family if their family member died during the study! However will it affect my results?

There is the obvious point that if a participant’s right to withdraw is hindered due to the incentive (as explained above) this could have a potential affect upon participants. For example if I was using a self-report method using a questionnaire, the participant may start to get bored or upset by the questions and normally they would withdraw from the study, however due to the cash incentive they feel obliged to finish the study. This would cause them to rush the questions and not put down truthful answers. This would decrease my internal validity of my results, as due to the false answers I am not measuring what I claim to be measuring, I appreciate that this is an extreme case, however I am using it to describe how cash incentives could affect psychological research.

On the other hand not being able to gain enough participants to be able to have a large enough sample size to make any generalizations easier, can be disastrous to psychological research. Occasionally without an incentive (of any kind) researchers might not be able to gain a decent sample size thus resulting in lower population validity. When a researches results have a low validity they are often overlooked as without valid results researchers have little power when backing up their results.

In conclusion I feel that incentives to take part in a study shouldn’t be used, due it potentially breaking some ethical guidelines, and also because of the way it affects results. However if there was a case when I couldn’t gain enough participants, then I would highly consider using an incentive of some sort.

---

[i] http://etext.lib.virginia.edu/healthsci/reed/commission.html#vol

Blog Entree Week 2 – Internal vs External Validity

External and internal validity are exceptionally important in psychology as they can show how generalizable the results of a psychological study are. When assessing results researchers should always use validity methods, as they, for example can then say whether their results are generalizable to the real world, and thus could improve everyday life.

External Validity consists of two main parts, population and ecological validity. Population validity is whether the results of a study can be generalised from the sample to the target population. So basically does the sample represent the people the study was aimed at. For example if a researchers study is too asses the motivation of Liverpool Football Club players, and the researcher had 50 Liverpool Football Club players of a range of ages and backgrounds, then you could argue that this study has a high population validity as his sample is very representative of his target population. On the other hand ecological validity is whether the results can be generalised into a real life setting. For example when a lab experiment is used, as the study is in an artificial setting, I would argue that the study has a low ecological validity and it is unclear the participants would act the same as they do in an articficial setting as they would in a natural setting.

Internal Validity is another useful tool for a researcher to assess his results. In simple terms Internal Validity means that if internal validity is considered as high then the researcher is measuring what he claims to be measuring. For example if the researcher is using a self-report measure and collecting subjective data about an embarrassing subject, then due to demand characteristics the participant might give false answers and thus the research has invalid research because he is therefore not measuring what he is claiming to measure. Maguire et al (2000) used an MRI scanner to measure the size of the anterior and posterior hippocampi located in the medial temporal lobe of the brain. Maguire wanted to measure the plasticity of the brain in regards to the memory of taxi drivers in London. As he was using an objective measure (MRI scanner) there is no way that Maguire was not measuring what he claimed to be measuring therefore his study would have a high internal validity.

As for the argument as to which validity is more important, personally I would say that Internal validity is much more important as if a researcher is not measuring what he thinks he is measuring then all his results are invalid and the study’s conclusions will not have sufficient evidence to back them up. Without a high internal validity, assessing the external validity is pointless as the results of the study are invalid anyway. However I can see how some people would argue that external validity is more important as they would argue that a study’s results are pointless if they cannot be applied to the real world, which is true and I also believe this. But as I have just stated if a study does not have a high internal validity then the results are corrupt and generalising corrupt results to the real world is pointless and would have a positive effect upon everyday life.