James Hartley argues that new large-scale research on titles doesn’t tell us much more than we already know. Effective titles attract and inform readers, and do this in a variety of different ways.


A colleague in my department thinks that if he uses a colon in the title of an article he is writing then the article will be accepted more readily, and cited more frequently, than if he does not. Where has he got this notion from?  A once carefully qualified statement in some research paper has now become a blind article of faith. True enough there is research on the effectiveness of such colonic titles, but it is not that convincing. But rumour and hearsay are powerful things.

Research on titles
The research on titles can be summarised as follows:

  • Almost all of the research takes place in the science and medical fields
  • Most of it was done before the advent of electronic counting
  • There is research on the effectiveness of grammatical features, such as colons question marks, the results of which I summarise below, and
  • There is also research on different types of titles, and on which kinds of title readers prefer.

More recent research
More recent research on titles, using electronic databases, has the benefit of huge sample sixes, but the findings are just as contradictory.  Consider the following studies:

Authors Nos. of Titles Subject Matter Main Findings
Buter & Van Raan (2011) 642,807 Web of science journals Hyphens and colons are common in some disciplines – and not including them correlates negatively with impact. Best to stick to disciplinary practice.
Jamali & Nikzad (2011) 2,172 Medicine & Life Sciences Titles with questions are downloaded more but cited less. Titles with colons are longer than titles in other formats but cited less.
Habibzadeh & Yadollhie (2010) 9,031 Science Articles with longer titles are cited more.
Ball (2009) C20 Million Medicine, Physics, Life Sciences There has been a huge increase in the number of titles with question marks.
Lewison & Hartley (2005) 216,300 science and 133,200 medical papers Sciences and Medicine Lengths of titles vary according to individual sciences: Biology averaged 15 words, mathematics 8. Colons present in 23 per cent of titles in medical sciences and 6 per cent in engineering and technology. Question marks rarely used. Single authors use colons more.

So, different authors prefer different types of titles in different disciplines, and there is little agreement about the effectiveness of these titles in terms of subsequent citation rates. The variations in types of titles generally reflect disciplinary practice.

Different types of title
A different tack in research on the effectiveness of titles is to consider how many different kinds of titles there might be. Currently I distinguish between 13 (see below) but there must be more. As yet I have no data on their frequency of use, and I am somewhat worried that some MA and PhD students around the world are now counting my title-types.

Different types of title 

  • Titles that announce the topic in general: The age of adolescence
  • Titles that particularise a specific theme following a general heading: Pre-writing: The relation between thinking and feeling
  • Titles that indicate the controlling question: Is academic writing masculine?
  • Titles that state the findings of a study: Asthma in schoolchildren is greater in schools close to animal feeding operations
  • Titles that indicate that an answer to the question will be revealed: Abstracts, introductions and discussions: How far do they differ in style?
  • Titles that announce the direction of the author’s argument: Plus ca change…Gender preferences for academic disciplines
  • Titles that emphasise the methodology used in the research: Using colons in titles: a meta-analytic review
  • Titles that suggest guidelines and/or comparisons: Seven types of ambiguity
  • Titles that bid for attention by startling openings: ‘Do you ride an elephant and never tell them you’re German’: The experiences of British, Asian, black and overseas student teachers in the UK
  • Titles that attract by alliteration: Legalese and legal ease
  • Titles that attract by using literary or biblical allusions: Lo! They came to pass. The motivations of failing students.
  • Titles that attract by using puns: Now take this PIL (Patient Information Leaflet)
  • Titles that mystify: How do you know you’ve alternated?

So what are the important questions?
I used to think that the purpose of titles was, in equal measure, to attract readers and to inform them about a paper’s content. I now think that the latter is more important.  Today too many titles lack crucial information and are thus misleading. For example:

‘Evaluating research: from informed peer review to bibliometrics’ or, ‘Informed peer review and uninformed bibliometrics’.

These two papers seem to be covering much the same thing, but title 1 is about national research assessment exercises, like the REF, and title 2 is about assessing applicants for research grants. Including this information lengthens the titles, but they would be more informative.

Similarly, adding more information to title 3 leads to title 4 which, in my view, is more effective:
Original title:  ‘University students’ estimates of occupational intelligence’
Revised title:   ‘How intelligent do you need to be to be a Surgeon?  Men and women students’ estimates of the intelligence required to carry out male, female and gender-neutral occupations’

Writing the title to an article is not easy. A lot depends upon where we are sending it.  Different disciplines – and sub-disciplines within them – have different styles. Blogs, too, require more dramatic, eye-catching titles. In my view it is best to work with a working title, and then decide on the actual title when completing the piece. My colleague may use a colon if it is appropriate – here I managed a colon and a question mark! 



Print Friendly