Writing a Research/Project Proposal

This is my personal opinion on how to write certain sections of a project including research and final-year projects (FYP) proposal. The following are sections that are typically required in a project proposal. Note that not all proposals strictly follow the structure described in this article. In Malaysia, however, when writing proposals for a Master/PhD research work, a final-year project, a research grant, the proposal's structure is similar to the one described here.

Abstract

According to the Oxford Dictionary, an abstract is a "summary of the contents of a book, article or speech". Therefore, an abstract should contain a summary of relevant sections in your project proposal. For a proposal, you may begin with a short introduction, then to the research problems, objectives (what you plan to solve) and methodology (how you plan to solve the problems). The final report or thesis should include a brief mention about what you have accomplished (e.g. the solutions/findings which have addressed all the research objectives).

Introduction

The introduction section provides an overview of the proposed project. Relevant works that lead to you to do the project should be briefly discussed in this section. Detailed discussion should be included in the Literature Review section (normally a separate section after the Introduction section). The content of this section typically includes the Research Statement and Objectives (both described below).

Problem Statement

Based on the literature review, highlight the problems that have been identified. The problem statement should be concise and not include any citation. Details of the problems should be explained at length in the Literature Review section. This section can be seen as a summary of the problems that you have described in the Literature Review section. After you have stated the problems, provide potential methods or solutions to solve them.

For an FYP, you may want to improve an existing system. Therefore, you can highlight the problems of the existing system as the problem statement.

Objectives

The objectives must address the problems that you have stated in the problem statement. We should be able to map the objectives to the research problems. One research problem may be addressed in one or more objectives.

• The objective typically begin with the word "to". For instance, "to develop ...", "to improve ...", "to allow".
  
• The objectives should be listed numerically (e.g. 1, 2, ...) and not lumped together in one paragraph or written as separate paragraphs.
• The objective must be measurable. If one of the objectives is to develop an algorithm, then at the end of the project, you must explain about the newly developed algorithm in the final report or thesis. If one of the objectives is to create a face-recognition functionality, then you must be able to provide a demonstration of this feature and to include details of its implementation in the final report.

Examples of Objectives

1. To allow registered users to login using their face.
2. To develop an algorithm that is able to perform online authenticated encryption.

For the first objective (perhaps relevant to a final-year project), at the end of the project, you must be able to demonstrate such functionality. For the second objective, at the end of the project you must be able to describe an algorithm that is able to provide such functionality.

Conducting Literature Review as One of the Objectives?

I have seen many proposals that include conducting literature review as one of the research objectives. Ideally, conducting literature review should not be an objective because it is an activity that you must perform before you prepare your research or project proposal. However, the absence of doing a literature review in the objective does not mean that you no longer need to further refine your literature review. Reviewing the literature is a continuous process that should be done throughout the project since new research may pop up along the way, or we may miss several works during our initial literature review.

Research Questions

Certain grant applications or research proposal require you to come up with research questions. The research questions must be derived from your research/project objectives. One objective may have more than one research questions.

Literature Review

A literature review should not just report all possible research or works that have been done before. You may start with a general introduction to your topic, and then focus on works that are relevant to your topic. Synthesise the literature by doing comparing relevant works, highlighting the advantages and disadvantages, their similarities, etc. Describe the problems/gaps that you have identified in the literature which you plan to solve in your project.

Include Only the Most Recent Works (e.g. past 5 years)?

I once encountered a reviewer who was adamant that works in the literature review should only include recent works (e.g. past 5 years) and everything older than that should be omitted. There was also a reviewer that insisted that the majority of references should consist of recent works (e.g. past 5 years). 

I do not quite agree with these statements. As long as the research problems are valid and that you have extensively gone through recent works and include them in your literature review, then it is OK to only have a handful of references to recent works. It may be the case that researchers are looking at other problems, or the problem is difficult. You should also be confident that you may be able to provide a better solution than existing ones to the research problems that you have identified.

In the cryptography field, an example of a solution that came long after the research problem was put forward is identity-based encryption scheme. The idea was proposed by Adi Shamir in 1984 and the first practical solution to the problem was proposed by Dan Boneh and Matthew Franklin in 2001.

Methodology

The Methodology section should explain how you plan to conduct your research or project. For certain projects, you may follow an existing methodology such as Agile or the Waterfall model. You may also create your own plan such as dividing the project into several phases. The methodology may also include the tools that you are planning to use in the project.