PDA

View Full Version : Advice; how to conduct a scientific study?


andyandy
20th August 2007, 08:38 AM
this stems from my thread in RP - "should religious education be in schools" (http://forums.randi.org/showthread.php?t=90626)

In which I argue that religious education should be in schools. If you wish to argue against this, please see this other thread :D

but in this thread i am genuinely interested in how one joins up the dots in actually realizing a scientific study which seeks to measure the effects of RE upon children's opinions and prejudices. These studies have been conducted on a number of other related educational topics, and have been shown to have significant success

Learning about each other’s differences is the way to reduce prejudice

Professor Rupert Brown, together with Professor Miles Hewstone of the University of Oxford [conducted a meta-analysis of studies in which they] tested their theory with experiments and field studies in settings as wide-ranging as Catholic-Protestant relations in Northern Ireland, attitudes between different nationalities in Europe, Muslims and Hindus in Bangladesh, and young children's attitudes towards peers with disabilities and other stigmatised groups. In a very recent study with primary school pupils in Kent, conducted with Lindsey Cameron and Adam Rutland (from the University of Kent), groups of children were presented with different types of stories involving an English child and a refugee child, working together on various projects. In some of the stories, only the characters' names and individual characteristics were given. In others, the emphasis was continually on the fact that they belonged to the same school; in a third group more detail was given about the characters' different cultural backgrounds in their common school setting. After six weeks, the researchers found that the group given stories that contained the cultural references brought about the most positive change in the children's attitude towards refugees.

Professor Brown explains: "In multi-ethnic schools, where conflicts can sometimes arise, it is beneficial to integrate the students and to get them working collaboratively. They then begin to regard each other not just as individuals, but also as people who are Afro-Caribbean, Asian or White. What needs to happen is that they are reminded about each other's cultural background, so that they can make a connection between the peers they have met and all others from similar backgrounds." He adds: "These types of beneficial intergroup contact cannot resolve all the issues of intergroup conflict, but can make a real contribution to its reduction". http://www.sussex.ac.uk/press_office...media482.shtml

for a study on RE, the hypothesis could be religious education in children can reduce prejudice against atheists,
and as an example
60 children could be suitably chosen to be split into two groups,
both groups spending 1 hour a week receiving education spread over 2 1/2 months (10 hours)
one group (control) receiving non religious specific education - say an English lit course
the second group receiving religious specific education based on an existing national RE syllabus (eg UK) which seeks to promote understanding of a wide range of religious belief
children to be of the same school age - say age 12-13 and the syllabus material to be reflective of key stage 3 level (UK) lessons.
both groups of children to fill in a detailed questionnaire both before the start of the course and upon its completion designed to measure their opinions on a wide range of faith related issues
A follow up questionnaire to be completed at a designated interval in the future (6 months/1 yr)

that's just off the top of my head as a rough outline. So what would the protocol be if (hypothetically speaking) i wished to see such a study conducted? What professionals would be most qualified to conduct such a study, would it have to be affiliated with an academic institution, can interested groups finance such study, how much (roughly) would such a study cost?

many questions.

cheers :)

andy

wollery
20th August 2007, 09:16 AM
The protocol sounds good, although it's not my field of expertise.

I think the best place to conduct such a study would be somewhere like the University of London Institute of Education (http://ioewebserver.ioe.ac.uk/ioe/index.html)

Dancing David
20th August 2007, 09:27 AM
There are many ways to do it. Sociological surveys and observation would give you the power of numbers. Longitudinal and demographic studies would give more complete data on complex interactions.

The difficulty is that you can't really have a double blind, it would take large group longitudinal studies to get an accurate sample and you would have to randomise the groups for huge amounts of diversity in the demographics.

andyandy
20th August 2007, 01:02 PM
thanks for the replies...

i'm going to see if i can take a bit of a break from JREF, i have been posting rather a lot recently :D

but any other comments gratefully received, and i'll be reading and responding to them all in due course :)