Monthly Archives: January 2011

Metadata>paradata. Discuss.

Today’s S3RI methodology seminar was given by Gabi Durrant and Julia D’Arrigo.  The subject was paradata, which I’m pretty sure is data which describes the process of survey data collection.  It might be data about the person conducting the interviews, or … Continue reading

Posted in Academic research | Tagged , | Comments Off on Metadata>paradata. Discuss.

Values and conflict in the voluntary sector

Malin Arvidson, who is part of the Third Sector Research Centre, gave a seminar yesterday about some of the work she has been doing on the Real Times project.  The Real Times project is longitudinal (which means that it follows people … Continue reading

Posted in Academic research, Methods | Tagged , , | Comments Off on Values and conflict in the voluntary sector

Time-varying covariates III

A thought occurs.  Longitudinal data allows change to be studied within individuals was well as within populations.  This is ‘a good thing’.  You can look at whether a training programme affected an individual’s employment status, for example.  Or whether breakfast … Continue reading

Posted in General | Tagged , , | Comments Off on Time-varying covariates III

Population projection – and it’s depressing

My take-home message from today was slightly depressing.  Sure, women do indeed live longer than men – but they live nearly all of those extra years with a disability.  Sigh…  I have a lingering interest in long-term care (for which … Continue reading

Posted in Academic research, Demography, Methods | Tagged , , , | Comments Off on Population projection – and it’s depressing

Scaremongering journalist…? Not this time.

I’ve just been reading a Guardian article about a BMJ article about exclusive breastfeeding for six months.  As a member of a university, I’m privileged to have access to many journals online, including the BMJ.  I used that privilege to … Continue reading

Posted in Academic research, Journalism | Tagged , , , , | 1 Comment

Time-varying covariates II

My division runs a reading group.  The subject is always methodological – this semester it’s a textbook on longitudinal data analysis (“Applied Longitudinal Data Analysis: Modeling Change and Event Occurrence” by Singer and Willett, for those who are interested).  The … Continue reading

Posted in Methods | Tagged , , | Comments Off on Time-varying covariates II

Bandwagon vs. threshold

I saw an interesting presentation by Helen Margetts from the Oxford Internet Institute at Southampton last month.  Her research suggests a threshold effect for political participation.  Under her experimental conditions, people were statistically significantly more likely to participate politically (in this case, … Continue reading

Posted in Academic research | Tagged | Comments Off on Bandwagon vs. threshold

Great post from the Harvard blog

Matt Blackwell at the Harvard Social Science Statistics Blog draws attention (in a most amusing way) to this interesting NY Times article.  Worth a read.  Basically, a professor with “a great sense of humour” (Daryl J. Bem) claims to have demonstrated … Continue reading

Posted in Journalism, Other blogs | Tagged , , , | Comments Off on Great post from the Harvard blog

Time-varying covariates

In an analysis of longitudinal data, you find variables which change over time (e.g. marital status) and those that do not (e.g. date of birth).  I was forced to think about the place of time-varying covariates in longitudinal data analysis … Continue reading

Posted in Academic research | Tagged , , | Comments Off on Time-varying covariates

Founding the Confounding Factor

What is a confounding factor? According to the Oxford English Dictionary, confounding is “destroying, confusing, perplexing, amazing, etc”.  In statistics, a confounding factor or variable worms its way between your carefully chosen cause and effect, making a nonsense of your research … Continue reading

Posted in General | Tagged | Comments Off on Founding the Confounding Factor