Writing Research Papers Writing Successful Grant Proposals Prof Alistair Sutcliffe Universities of Lancaster Manchester

Writing Research Papers Writing Successful Grant Proposals Prof Alistair Sutcliffe Universities of Lancaster Manchester Presentation Outline 1 Preparing a Research Paper 2 Styles Structure 3 Gen.Writing Research Papers Writing Successful Grant Proposals Prof Alistair Sutcliffe Universities of Lancaster Manchester Presentation Outline 1 Preparing a Research Paper 2 Styles Structure 3 Gen.

Writing Research Papers

&

Writing Successful Grant Proposals

Prof Alistair Sutcliffe

Universities of Lancaster

& Manchester

Presentation Outline

1.  Preparing a Research Paper

2.  Styles & Structure

3.  General Advice

Phases of a research study

–  Decide on specific procedures/methods to

answer our question

•  Testing phase

–  Use the procedures devised in the previous

phase to test your idea/theory/model

•  Interpretation phase

–  Answer research question based on results

from testing

•   Communication phase – the research paper

–   Should include a description of all preceding

phases

Enabled & supported

by a review

of literature

Writing Research papers

1 Planning & Repartition

Planning Research Papers

•  What is the “story” ?

- this should be set by your Research questions and

- the results

•  Why is it interesting ?

how does it relate to previous work ?

•  What is the contribution ?

- type of research

- how has it advanced the state of the art ?

Journal or Conference ?

•  Quantity and Quality of the research

substantial study taken 1-2 years > Journal more limited study this year -à conference

•  Select which journal/ conference depending on your ambition and the quality of the work

top quality conferences acceptance rate 10- 15% medium quality conferences 20-25 %

poor conferences 50% +

Journals

•  Report significant research

•  Most important source of results & reputable research

•  Journal articles give more detail on previous research, methods and results

•  Prefer refereed journals

–  Peer review: academics unknown to the author assess the quality of a paper before publication

•  http://www.elsevier.com/wps/find/

journaldescription.cws_home/642704/authorinstructions

Journal papers

•  Acceptance rates depend on journal – but usually

20-50%

•  Process is slow- depends on reviewer response time

-check journal web site for statistics

•  Acceptance depends on

- choosing the appropriate journal

- quality of the work

- answering the reviewers’ comments

Journal Paper Review

1.  Submit paper

2.  Editor/ Associate editor assigns reviewers

3.  Reviews considered by Associate Editor

4.  Receive reviews and decision: Reject/Major-minor

revisions

5.  Edit- revise paper

6.  2nd version sent to same reviewers

7.  Receive reviews and decision: Major-minor revisions/

Accept

8.  Edit/Prepare final version

9.  Paper accepted and published

Whole process 12-24 months

–  Many other series

•  Proceedings- check availability- Digital libraries

Conference Review Process

1.  Submit paper

2.  Program Committee considers reviews (usually 2-3)

3.  Receive reviews and Accept/Reject

4.  Answer Reviewers points (rebuttal)

5.  Program committee makes final decision

6.  Edit paper, published in proceedings

(whole process 4-6 months)

Which Journal / Conference ?

•  You should know the answer from your awareness of the literature, contact with colleagues, etc

•  Conferences- International rather than regional, ACM, IEEE, IFIP in the title

•  Journals- International rather than regional ACM / IEEE transactions

•  Look at lead times for publication and impact factors (journals) >1 good for computer science NB Medical

journals tend to be >4 !

Which Journal- Conference ?

Your audience/ community ?

•  Do you know the journal conference/ journal, people on the program committee/ editorial board ?

•  If no,t then check out who they are and what style of research they do

•  Style- qualitative v quantitative, formal v informal,

theoretical v practical-applied

•  If you are unsure- ask the editor (journal), look at

previous proceedings + CFP for the conference

IT/Computer Science Conferences

ACM Assets (Assistive Technology), ICT4D (IT for

developing countries), ECIS (information systems) and a host of other specialist areas

3rd Rank

HCI International, WESAS, and many others- tend to

accept extended abstracts

ICT/Computer Science Journals

1st Rank

IEEE and ACM Transactions- TOCHI, TSE, TOIS

MIS Quarterly, ISJ, HCI Journal

•  Books

–  Monographs focused to an academic audience, quality depends

on author and publishing house

•  Book chapters

- Report results, frameworks, reviews

- quality depends on editor and author

•  Books provide a good for an overview of the subject

2 Research/ Paper

Styles &

Structure

Paper (research) Styles

•  System development

- Use and integrate established techniques in order to

build a computer-based system

•  Empirical research

- Investigation of an IT related problem and/or

development and evaluation of a computer-based

system

•  Applied project

- Anchored by industrial/ real world problem, may be software development or case study

Paper Styles Analysis Investigation- quantitative

Abstract

1 Introduction

2 Critical review of relevant research

3 Research methods- Experimental Design

4 Data collection and statistical analysis

5 Results

6 Conclusions, Limitations

References

Appendices (questionnaires, descriptive statistics)

Methods- Questionnaires, Experimental observations

Appendeces (interview transcripts, case study details)

Methods: Questionnaires, interviews, observation, ethnography

Software Development Papers

Abstract

1 Introduction

2 Critical review of relevant research

3 Requirements/Specification/Model

4 Design and Implementation

5 Testing and evaluation

6 Conclusions

References

Appendices (code, detailed specification, test results)

3 General Advice

Literature review

&

Writing Style

Literature review

•  The main reasons for reviewing literature:

–  Preliminary search which helps generate & refine research

ideas

–  Establish the baseline for your research- make sure you do not

‘reinvent the wheel’

–  Necessary for the ‘related work’ section of a paper

–  More important in Journal papers

–  Forms the basis for the discussion in your paper

–  Critical review - establish what research has been published in your chosen area & try to identify other research currently in progress

Literature review- questions

•  What are the main questions that have been

investigated in the field?

•  What are the main theories about the topic?

•  What are the key sources?

•  What are the main debates about the topic?

•  What are the main concepts and constructs within the topic?

•  What evidence is lacking, contradictory, inconclusive or limited?

•  Why is it still important to investigate the topic?

Planning the search

•  Key words are basic terms which describe your

research question(s) & objectives

•  Generating Key Words

–  Define your topic in one phrase

–  Split the topic into different concepts

–  Find alternative terms for the concepts

–  Author names

•  Reading articles by key authors & recent review articles

in your research area helps to define your subject

matter & suggest appropriate key words

Sorting the wheat (good stuff) from the chaff

(rubbish)

•  Look for publications by authorities in the field, keynote speakers at conferences, people who other people cite in references, on the editorial board of journals etc

•  Check the quality of the source- journal Impact factors, paper/

author citations indexes, e.g

–  Citeseer http://citeseer.ist.psu.edu/

–  DBLP http://www.informatik.uni-trier.de/~ley/db/

–  Google Scholar (not up to date)

•  Use web site authority, e.g Scientific Web to search

•  Check the authority of the site author/owner- see previous

guidelines and reputable University ? Reputable organisation/

company ?

•  Follow links from reputable sites and don’t trust everything you find via Google

Good vs bad literature reviews

breadth and depth

–  has clarity and

–  provides only descriptions –  is narrow and shallow –  confusing and

longwinded –  is constructed in an arbitrary way

Reference the literature

•  Purpose of referencing is to enable the reader of your paper to locate the actual material which you are

referring to, quoting or acknowledging

•  Match the citation (marker) in your text to the entry in the complete list at the end

•  Explain what the paper actually reports/ claims

•  Motivating literature, comparison with your work

Scientific Writing- Style

•  Avoid 1st person- no ‘We’, ‘I’, third person or passive voice

–  “the study was conducted between May and August 2008”, rather than “I did my research over the summer”

•  Generally use the past tense

–  “the results of the experiment demonstrated ”

•  Use more formal language, no slang

•  Explain acronyms and avoid jargon

Scientific Writing- illustrations

•  Illustrations

–  clarify explanation

–  offer a different viewpoint

–  save words

•  Use graphs, photos, diagrams, maps, tables, where they are appropriate

•  Don’t duplicate text and illustrations

–  e.g present a table of complete results, but in the text simply draw attention to the significant ones

•  Reference the illustrations, e.g

–  “Table 2.1 shows …”

–  “… (see figure 3.4) …”

•  Be professional

–  use a consistent style

–  get help if necessary

Summary

•  Plan the ‘message’

•  Outline- plan the structure- theme

•  Report concisely and illustrate

•  Discuss the contributions

•  Choose your audience

Writing Sucessful Grant Proposals

Prof Alistair Sutcliffe

Universities of Lancaster

& Manchester

1 Planning & Preparation

Planning Research Proposals

•  What is the “problem” ?

- this should be in the Research questions and objectives

•  Why is it interesting ?

- how does it relate to previous work ?

- why is it novel, original, timely, relevant ?

•  What is the contribution ?

- type of research

- how will it advanced the state of the art ?

Pilot studies

•  Have you tried out your research idea in a test project ?

- Master student projects, PhD student research

- Do some of the work yourself

- Ask your University for funds to do a pilot study

•  Pilot projects help because:

–  you have some evidence that the idea might work –  you have thought through the problem

•  Test your ideas with colleagues, in practice

•  Literature Survey

Has your idea already been done ?

Has something similar been done ?

What research are you building on ?

•  Previous work- related research

Justifies your work- you are developing ideas from {x} Establishes why your proposal is unique – different from {a,b,c,}

Justifies why your work is important-

- scale of the problem

- theoretical need,

- applied socio-economic need

Previous work

Call for Proposals

•  Research programs can be specific with a Call or general Open competition

•  If you are addressing a Call for Proposals

- Read the call carefully

- Is your proposal relevant, why ?

•  Look at the criteria for proposals

- Can you match them ?

- How popular witll the call be (trendy topics)

- Can you compete ?

2 Writing the Proposal

The Idea

•  There is no substitute for a good idea

•  A good research proposal should be

Novel Timely

Research Plan

•  How will you address the problem ?

•  What type of research will you carry out to solve the

problem/ develop the idea ?

{see Research Methods workshop}

•  Do you have the skills and experience in this type of

Proposal Structure

(see you local funding organisation for detail)

1.  The problem, justification for the research

2.  Objectives and Research Questions

3.  Approach- how you will address the problem/idea

4.  Methods and workplan- detail on how you will address

the problem, tasks, outcomes, deliverables

5.  Project Management, Bar Chart, risks

6.  Your track record, expertise

An Example- SAMS project

(UK EPSRC Applied Research)

1.  The problem & why important ?

•  Many Old People get Dementia (600,000 + in UK)

•  Most need full time care £ 50 /annum

•  Most people get diagnosed too late

•  Early diagnosis can delay the disease with drugs

SAMS Project

Research Questions

•  How to convert clinical expertise into monitoring rules

to interpret patterns of human behaviour through

human-computer interaction and text messages

•  How to combine data and text mining to infer clinically

meaningful changes in level of cognitive function,

including the avoidance of false positives

•  How to deliver user feedback that is not only effective

but also socially acceptable, sensitive and

empowering to aid future decision making

•  How to deliver an effective system in an

ecological-social setting, a process problem of participatory

co-design with older volunteer users

SAMS Approach

Computer Science Research

1. Requirements- how will people react, what monitoring is acceptable ?

2. Modelling- linking clinical expertise for diagnosis to

computer measures of user activity

3. Design- specification of Data and Text Mining process to analyse user activity

4. Development of Prototype system

5. Evaluation and Testing with users over 12 months

- will the software actually detect the the difference between people who have early dementia and

those who don’t ?

- will be system be accepted by people ?

SAMS workplan

T1 Requirements and participatory design (lead PS,

with AB, CB, IL, RA2+4)

T2 Specification of detection processes

T3 Data mining technology

T4 Text-mining technology

T5 Inference & integration

T6 Evaluation and validation

T7 Generic Architecture delivery and assessment

Beneficiaries

National Health Service, Older People

Medical Technology companies

•  Correct ?

- Have you covered all the sections in the proposal form ?

•  Clear ?

- Is it easy to read and concise ?

- review proposal with experienced colleagues

and Finally

•  Success rates are low 10- 20 %

•  If you succeed- congratulations, make sure you deliver

•  If not, Don’t give up !

- read the reviewers’ comments, improve your

proposal

- try a different funding agency/ call