The average adult has one ovary and one testicle.

I enjoyed Knight’s warning on pg 179 about statistical significance. Proving something is statisically true does not neccessarily also prove it is socially relevant. I took a stats course before, and that prof used the statement in the title of this post to help illustrate that to us. It was helpful to me and quite fun, so I thought I’d share with you lot.

In reality, though, when you’re doing real research, this is a real tricky thing. I mean, how do you as the researcher decide which statisical truth is relevant and which is not? How do you keep your own bias out of it?

Hannah

Advertisements

2 thoughts on “The average adult has one ovary and one testicle.

  1. jessstarr88

    First of all – Hannah – I really like the title of your post. I’ve never heard that before and it was actually really helpful in demonstrating some of Knight’s points on Numbers and Data Analysis. I have been thinking about numerical data since writing the SSHRC application, and frankly it has been stressing me out a bit. It was a relief for Knight to write that in social science research numerical data was not necessarily the most important thing, but I’m still having a hard time wrapping my head around it.

    The examples that Knight provides while discussing statistical significance are helpful – especially the one on education experiences in schools (180), and I think I understand the challenges researchers face in determining how important these numerical relationships are. However, I’m still having a hard time imagining HOW I would actually turn a survey, interview, or other qualitative study, into numbers. I have looked at the “Are virtual reference services color blind?” article from this week, and see that this is an example, but (embarrassingly) I have not really read this type of report before, and I think it would be helpful if I could see an example of this that’s a bit simpler. If any of you have come across such an example of numerical data in qualitative research for dummies – I would really appreciate it if you could send it my way!

    Jessica

    Reply
  2. bradleymcilwain

    Hi Hanna,

    I have also found myself, while reading Knight, struggling with the question of how do we guard ourselves from bias, by both the researcher’s perspective and the subject being sampled. This is where triangulation, or what Knight calls multi-method design, I believe can be a useful tool for researchers by secondary sampling, using the same or different methods, as a way of fact checking or assessing a sample’s reliability to protect against bias. Knight suggests, however:

    “if social phenomena are coloured by context, if chameleon-like they change, if observers legitimately see differently, then triangulation should produce more complex accounts and it is likely that there will be disagreements,” (Knight: 127) and also states that “this still leaves open questions about how researchers can deal with findings that cannot be neatly lined up to point in the same direction” (Knight 128).

    I disagree with Knight. I feel that a) it is because social phenomena are coloured by context and subject to change that researchers should be weary of and investigate the possibility of bias to make sure that information in the sample is reflected truthfully and accurately; b) the use of multi-method design may actually help substantiate samples and potentially clear up disagreements, instead of cause them – and the obvious question, why does Knight suggest that researchers should be worried that their complex accounts should cause disagreements? Should we, then, according to Knight, be conducting research that is simply in ‘agreement’ with ‘the literature’? c) research where ‘findings that cannot be neatly lined up to point in the same direction’ should be encouraged to avoid bias. How badly, as researchers, do we want to ‘line up’ our findings to prove our hypothesis, and, in doing so, what variables are we overlooking that could be important to our research?

    Bradley

    Reply

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