• Question: How do you test your theory?

    Asked by emmmma to Jamie, Jodie, Kat, Mark, Niamh on 15 Mar 2011 in Categories: .
    • Photo: Jamie Pringle

      Jamie Pringle answered on 14 Mar 2011:

      Hello again Emma,

      You are interested in this area, I like that!

      To give you an example, I am looking to see which main geophysical technique works best to find a murder victim. To do this, I have buried pigs as human proxies and then gone over them with magnetics, electrical and radar equipment. I then looked at the respective datasets and see which worked best!

    • Photo: Katherine Davies

      Katherine Davies answered on 15 Mar 2011:

      Wow Emma you have asked many questions!

      I test my methods using many repeats of the same thing. At the moment, I am examining the morphology/anatomy of fly pupae to see if we can use the changes in metamorphosis to determine age. I have examined over 1000 of them, to make sure I understand which features/developments I am correlating to each age of pupa. I also use blind-testing, where someone else gives me pupae that they know the age of, and I have to estimate it using my methods.

    • Photo: Niamh Nic Daeid

      Niamh Nic Daeid answered on 15 Mar 2011:

      Hi Emma – Theorys are tested by doing lots of experiments over and over. The reason for that is to gather as much data as possible to support (or otherwise) what you think is happening and so be able to provide evidence that your theory may be correct.

    • Photo: Mark Hill

      Mark Hill answered on 15 Mar 2011:

      Hi Emmmma, With all of your questions, your fingers must be getting tired as well.

      I generally don’t have a hypothesis, or hypotheses. I try to approach from a ‘grounded theory’ approach – looking at the evidence and seeing where it takes me. I start at the end, where the vehicles and people have come to rest, and try, through a reconstruction (not necessarily with the people, although dummies can be used) to work as far back along the time-line (chronology) of the collision event. The really important thing is to try and identify what caused the collision, and less importantly what happened at the collision point and afterwards, although the whole picture is important.

      Certain aspects of a collision investigation are sometimes tested. I investigated a crsah in which a six year old girl died, whilst sitting behind the driver in a small car. The driver mis-judged a right bend and drove partly onto the left verge. She braked and the car shot across the road, to the left, because the tyres on the right gripped the road under braking and the left tyres, on the grass didn’t hold very well. As the car crossed the road, it was hit in the left side by a van. The car then hit a tree head-on. At post mortem, we found that the girl had suffered a fatal fracture to her spine.

      I thought that, although she was wearing the correct belt on a booster seat, the side impact had caused her to slide to her left. She then couldn’t sit back up, in the diagonal belt, before hitting the tree. This caused her spine to twist and bend at the same time, snapping it. I measured her height, shoulder width and the seat. I then made a wooden ‘former’ to her size, cut the roof off of the car and placed the former on her seat. I was then able to see and photograph how she could have slipped and twisted. This was a really sad incident, but it showed that she was in the correct belt for her size, but the two collisions were so close together that she was unable to sit upright in time.

      I test theories by creating reconstructions and experiments to try and replicate collision situations.

      This is typically ‘me’ log winded, but I hope that it helps.

      Mark.

    • Photo: Jodie Dunnett

      Jodie Dunnett answered on 15 Mar 2011:

      I began by growing opium poppies in different soil types. I then analysed the different soil samples as well as samples of the leaves and resin. I made sure that I had lots of repeat samples so that I can perform a statistical analysis of my results. Once I have completed my analysis, I should be able to see if it is possible to identify where the opium was grown by the ratio of metal ions it contains.

Comments