Using Effect Sizes and Confidence Intervals with Correlation-based Analyses
Date:
27/11/2013
Organised by:
NCRM, University of Southampton
Presenter:
Dr Catherine Fritz,Dr Graham Smith, Dr Roz Collings, Prof. Chris Roe, Prof. Peter Morris and Mr Amar Cherchar - University of Northampton
Level:
Intermediate (some prior knowledge)
Contact:
Jacqui Thorp
Training & Bursary Administrator
National Centre for Research Methods
jmh6@soton.ac.uk
Map:
View in Google Maps (NN2 7AL)
Venue:
University of Northampton
Park Campus
Boughton Green Road
Northampton
Description:
Statistical practice and many textbooks aimed at social scientists have tended to strongly focus on null hypothesis significance testing, with little emphasis on the complementary information provided by effect sizes and confidence intervals. Recently, though, journals and funders have started to require the richer approach, drawing upon confidence intervals, effect sizes and power analyses. In addition, communication with non-scientists benefits from analyses and conclusions that do not rely entirely upon null hypothesis significance testing.
This hands-on, one-day workshop is designed for social scientists who conduct or read about research involving correlation-based analyses – i.e., correlations, regression, cross-tabs. We will address how to calculate effect sizes and confidence intervals, from raw data as well as from published statistics, and how to use this information to enrich interpretation of the data, enhance communication with stakeholders and improve subsequent research.
We ask participants to bring to the session (and send to us, in advance) a paper from their area of interest which includes consideration of correlation or regression. This paper does not need to be accompanied by the raw data.
Cost:
The fee is:
1. £30 - For UK registered postgraduate students
2. £60 – For staff at UK academic institutions, ESRC funded researchers and registered charity organisations
3. £220 - For all other participants
All fees include event materials, lunch, morning and afternoon tea. They do not include travel and accommodation costs.
Website and registration:
Region:
East Midlands
Keywords:
Analysis of official statistics, Confidence intervals , Effect size , power analysis , correlation , regression
Related publications and presentations:
Analysis of official statistics