Week 11: Research with Children March 28, 2013
- There are many ways to conduct research
- All research is embedded in culture and politics
- Assumptions about childhood shape the kind of research done on, or with, them
- There are some distinctions between conducting research on/with children and on/with adults
How do we know what we know?
- Our experiences
o Relying on this is not always the best way of understanding the social world
o Taking a more systematic approach to studying something
o Can involve studying lots of people
o Thought-through process, careful and follow process , explain process
Ways of Conducting Research
- Choose a strategy
o E.g. imagine you are studing bullying between students in elementary school: what do
what do you want to know? How will you find out?
To narrow down -> what is bullying? How are you going to define it? Are you
going to watch for it? Where? How? Are you going to ask about it? Who are you
going to ask? Where will you find them?
- Need to consider the following:
o Is there an objective world out there that we can access? (ontological question)
o Do you want to know what people do or what kinds of meanings they give to what they
o Do you want general information about a lot of people or detailed information about
o Are you seeking descriptive findings? Explanations?
o Quantitative and qualitative approaches
Quantitative Research Methods
- Can be calculated, involves numbers -> statistics
- Holds this idea of positivism
o Notion of truth (the truth is out there)
o Objective -> independent of our opinion
o Static – doesn’t change o Stable – information exists
o Categorization - can fit it into categories
- Employs methods into finding the truth
- E.g. Bullying example again
o Quantitative questions
How many times do you witness bullying in a week?
How many times have you been bullied?
How many times have you bullied?
What is bullying?
- Quantitative approach
o Collect “data”
Examples (what data might we collect?)
o Look for patterns
o Look to generalize results
o Should predict future behaviour
o Our methodology should be “replicable”
Example of Wakefield’s Autism study (autism and drugs)
No one could replicate that study, ever
Needs to be able to get around the same results -> use methods section to be
able to do the study someone else did (purpose of methods)
- Reliability is important -> bigger N, more reliability
o Statistical power
o A large N
Results are reliable
We tested what we thought we were testing and the results reliably
represent the phenomena we were studying
- Longitudinal research (repeated measures)
o Bullying example – interested in “development”
How might we study “bullying” using a longitudinal design?
Why would this be advantageous?
o School boards