When Statistics Ignores Morality.

Statistics Blog 1 (Sem 2)

After writing a research report using qualitative methodology in the form of IPA the aim of my blog this week is to discuss the importance of processing rich detailed data to make conclusions about humans, preserving their integrity. Rather than a straight comparison to the quantitative method I intend to use the book “The Bell Curve” (Richard J Herrnstein & Charles Murray) to form the argument of my entry.

Studies such as Zimbardo’s prison experiment and Milgram’s study of obedience have been well documented and explicitly taught, psychological harm and  physical harm to participants occurred during Zimbardo’s investigation and psychological harm could be argued in Milgram’s study. While harm on a singular occasion to a relatively small number of individuals (acting as participants) may seem dreadful to some and more than worthy of a student’s blog, I instead wish to discuss the harm that can come from the conclusions that studies using a purely quantitative method can arrive at.

One research report by no means makes me an expert, I did however learn a lot from the experience that has influenced my views. A qualitative study that builds rapport with participants during a conversation-like interview yields emotional content and fact that is specific to their own experiences, these experiences are often the same or similar to others, not because humans are so innately boring but rather because we all live in the same sorts of ways. Cross cultural studies will likely show different trends but they however would probably be matched by other members of their culture.

Quantitative methodology as I have stated in my last blog is a remarkable method for making inferences about cognition, examining reaction time and drug trials. However the latter has plenty of space for the qualitative method as often the best solutions are drug and methods such as community care. In the case of “The Bell Curve” it is so called due to the distribution we are all by now familiar with, the examination of IQ in relation to culture, race and wealth were explored with aims to advise the American government. Using mean scores of IQ for a range of racial backgrounds the two authors proposed that up to 80% of intelligence is inherited. They also used information on intelligence and cross referenced it with employment status, likelihood of incarceration and unwanted pregnancies etc.

Herrnstein died before the book was published and the information was ‘written off’ as anti-American and racist. Using the average of IQ scores to determine the worth of a race and to treat them differently, just because the average outcome of low IQ individuals seems undesirable to the cognitive elite is wrong. The information outlined in The Bell Curve may appear racist but to Charles Murray it was merely the report of statistics that were valid and the measures reliable.

While the data the authors gathered may have been true a more qualitative methodology would have allowed investigation of why such outcomes are prevalent in certain cultures and racial backgrounds. Wealth directly affects housing and therefor areas for schooling, it may well be that the race in question would easily have a higher IQ average if they were given the same opportunities, denying equal opportunity in capitalist countries is common, but to then drag their identities through the mud for the discrimination placed on them is wholly unfair to do let alone publish.

To conclude my rant, quantitative methodology is an excellent tool but should be used with care. Adopting a qualitative aspect can help appreciate the facts responsible for outcomes rather than an oversimplified and inaccurate singular cause and response ideal.

Advertisements

9 responses to “When Statistics Ignores Morality.

  1. Your blog highlights one of the main problems with qualitative research and that qualitative research needs to be approached with care. Qualitative research is criticised and called unscietific. It is more difficult to prove the validity and reliability in this type of daa. This does not mean, however, that qualitative research is less reliable/valid; it is just harder to prove. Like you said qualitative research does generate rich data. I personally think qualitative research can be really beneficial as you get an indepth insight into peoples experiences and use them to try and improve them in the future. I can understand that people prefer having numbers to look at to help them visualise the data, therefore I think (when possible) a halfway house method, combining both aspects, could be the best of both worlds and please most people.
    I did a quick search on the disadvantages and benefits of qualitative methods and found two links that clearly explained these points as a brief reminder. Here they are:- http://generallythinking.com/answers/research-methods/what-are-the-advantages-of-qualitative-research/
    http://generallythinking.com/answers/research-methods/what-are-the-disadvantages-of-qualitative-research/

  2. Pingback: homework for my TA (for the 8th feb 2012) « kennedy92

  3. Granted there are problems with statistics in terms of ignoring morality. Here’s an immoral example. “If Hitler was for climate change who would have been correct” This sounds terrible and disgusting to even consider but breaking it down the statistics of this statement are sound… Hitler murdered around 6million people take this figure and lets say since then at least two generations have been born. That’s 6milliont cubed which is a massive number of people if this additionally number existed in the world today imagine how much worse our global warming situation would be. This is a logically formed mathematically based argument that draws a totally immoral conclusion. However if psychology wishes to be a science morality should not be an aim of it. Science is about creating a conclusion based on empirical findings what ever that conclusion may be, whether it is moral or a-moral it is science. There should not be a fear of publishing papers because it has a result that shows somebody better than everybody else. I agree that at the point of inference it can become a-moral but that only because of human interpretation. On the other hand 100% of qualitative research is interpretation no matter how fastidiously you conduct that research. Qualitative methods can be equally as harmful we need only to think back to Dr. John Money (http://www.youtube.com/watch?v=MUTcwqR4Q4Y ) to remind ourselves how wrong the psychology can go even when using both approaches with a fairly good use of qualitative research. Our interpretation in both will always be the crux of the issue and where both can go awry and thus both methods have the same issue when it comes to conclusions and both need proper regulation in order to be carefully monitored to prevent the conclusions of these studies harming people. If anything qualitative research has more chance of going morally astray.

    • Quantitative methods suffer their ability to be extrapolated, losing sight of individuals treating poulations by average losing the identity of individuals and stereotyping groups my means. Self fufilling prophecies are seen to show some damage, in terms of moral violation both approaches have dark pasts (little Albert etc). Each approach should be left to what it fits best, so when advising social protocol quantitative methods shouldnt be involved, as qualitative shouldnt be involved in reaction time analysis, both have the potential for moral harm/violation when used improperly. I must however agree with your example demonstrating the flaws within qualitative methodology as the area has many, like the quantitative method. However i disagree with the assertion that all scientific findings, even when methodologically sound, should be published regardless of content. Like free speech surely there is a tipping point for psychological papers where the statement of fact turns to alienation and discrimination, shaping a cultural atmosphere simmilar to that of latin america where the poor live literally bricked apart from the rich and the rift between the two becomes irreversible and harmful. This type of scenario is what The Bell Curve warned could happen in america and other developed countries. Just becuase something is fact and proven to be so doesnt warrant publishing, even if it were to just be published for the sake of scientific merit it doesnt take much, if any imagination, to forsee a time when a person uses it to lead a political shift in segregating the elite. The Conservative party would be a prime candidate as the work (The Bell Curve) is one of early neoconservatism, nepetism, discrimination and wealth the first order of business. in addition why publish work outlining why certain types of people are less intelligent,valuable or able than others why not research and publish work assisting people not only on the lower end of the spectrum ,of intelligence etc, but also the higher end. You can curse the darkness or light a candle.

      These articles outline ‘lighting a candle’and you might find them intersting.
      Teaching skills for those with autism http://www.getcited.org/puba/101487277
      Discussion of methods to excel in already intelligent individuals http://www.jstor.org/pss/2090662

      The Bell curve:
      http://en.wikipedia.org/wiki/The_Bell_Curve
      http://www.bookdepository.co.uk/Bell-Curve-Richard-Herrnstein/9780684824291

  4. Pingback: TA Comment Homework Wednesday 8th Feb | cgpsychblog

  5. Pingback: Homework For my TA Blog comments 1 semester 2 « robinson8040

  6. I agree with this, to an extent. I read a portion of Ronald P Carver’s ‘The Case agains Statistical Significance testing’ (http://scholasticadministrator.typepad.com/thisweekineducation/files/the_case_against_statistical_significance_testing.pdf) and he explains how statistics can have a significant difference but that is incomprehensable and negligable. I find one problem is that you can have findings that are non-significant but by simply adding more participants, ie copying and pasting your data, to get a larger sample you can get statistical significance. As useful as this technique is surely, in a world where scientific publications are so important and highly regarged, this is wrong and lying?

  7. I think that the uses of quantitative and qualitative data combined are essential to truly understand experiments such as Milgram’s.
    If we only look at the quantitative data produced by Milgram’s research we would think that the world is crazy and people would happily electrocute people until they had reached the maximum number of volts whether the felt bad or that the “learner” was in serious pain.
    However, if we then look at the qualitative data, we can see that people weren’t actually very happy and most requested to stop the experiment. We also get to see the effects the experiment had on the participants and how much they wanted to stop but felt unable to because of the experimenter telling them that they must carry on.
    For experiments like this, I think that both methods are essential for a full understanding of the experiment, including exactly what the participants went through and why the quantitative results are what they are.
    In conclusion, while both methods have their advantages and disadvantages and experiments that one suits better than the other, I think that a halfway house is required for most experiments.

  8. Pingback: Homework for my TA Week 8 Comments « hls92

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s