91̽��

Introduction to the Research Cycle

Quantitative research involves the following stages:

1. Plan - Identify a research question
2. Data collection - Collect and organise the data
3. Data analysis - Calculate & Conclude
4. Reporting the results- Communicate & Display

Sometimes these stages are referred to as “The Statistical Cycle”. There are many ways of labelling and defining the different stages. Whichever definitions we use, broadly the same stages are involved.

There are more details on each of these stages below or at this

---

Stage 1: Plan - Identify a Research question

What to do

This is where you use your subject knowledge to identify what you’d like to research. You then need to turn your area of research into a clear, well-defined question. You need to form Null and Alternative Hypotheses and consider how the data might look depending on which Hypothesis is true.

This process might look something like this:

• Area of Research: Fashion trends
• Specific area of Interest: Trends in which colours are worn at different times of year
• Research question: Are blue clothes worn more in summer?
• Null Hypothesis: Blue clothes are worn the same amount in summer and at other times of year.
• Alternative Hypothesis: Blue clothes are worn more or less frequently in summer compared to other times of year.

Once we have reached this stage, we need to think about what data we need to collect in order to answer our research question. Before you collect your data, make sure you have decided what calculations you plan to do. To do this, the following questions are often useful:

• What data do you need?
• When do you need to collect your data?
• How will you collect your data?
• Why are you collecting your data?

What are the statistical processes I need to know about?

This stage of the cycle largely involves your knowledge from your area of research. However, it’s important to know about the statistics involved in the next stages of the cycle as that will help you form practical research questions.

Links & Resources to help

Statistical Analysis Planning workshop from MASH. This workshop runs at several points in the year and can be booked on or

---

Stage 2: Data collection

What to do

Make sure you know how you will analyse your data before you collect it. You should have a plan for what analysis you want to perform, what hypothesis tests you will perform and what calculations you want to do. If you don’t plan your analysis, you may find that you spend a lot of time and effort collecting data only to realise that you can’t use the data to draw any meaningful conclusions.

You then need to consider how you will collect or gather the data.

Will you collect the data yourself (primary data) or use existing sources of data (secondary data)?

For primary data, you need to consider how you will keep your sample unbiased, how large your sample needs to be and exactly what questions you will ask or what measurements you will take.

For secondary data, you need to consider what sources of data you will use and how reliable they are.

You should also consider how you will enter your data into a computer to perform your analysis. Plan ahead and think about what software you will use and how you will use it. When you enter your data into a computer, do so in a way that will make it easy to carry out the analysis that you have planned.

What are the statistical processes I need to know about?

Questionnaire design

Techniques for dealing with Missing data

Bias in data collection

Links & Resources to help

Data Collection methods

Survey Sampling Methods

Bias in Survey Sampling

---

Stage 3: Data Analysis - Calculate and Conclude

What to do

This is the part of the cycle which we think of most often when we think about statistics. This is where we carry out calculations with your planned analysis and interpret the results. A key element is to choose the right analysis or Hypothesis test to investigate the data. Next, we need to check that your data fit the assumptions for that test. You should decide what analysis to use in the planning stage before collecting your data.

The data analysis stage can be broken down into the following steps:

Step 1: Data cleaning 

You may also need to “clean” the data before you can perform an analysis eg. what will you do with missing values, outliers or errors in the data recording process.

Step 2: Descriptive summaries of the data

This involves summarising the main features of the data, including its central tendency (average), variability and distribution. The following summaries are typically used for the following types of data:

• Mean and the Standard Deviation for normally distributed continuous data
• Median and the Interquartile range for skewed continuous data
• The Minimum and Maximum is sometimes presented for continuous data in addition.

• The number and proportion for categorical data

Step 3: Analysis of the primary outcome/research question

This step involves using statistical tests and models to analyse your data. Remember  to check the model assumptions prior to conducting the analysis.

This step also involves displaying your data graphically, for example for continuous data you may use box plots or histograms, and for categorical data you may use bar charts or pie charts.

What are the statistical processes I need to know about?

This is where you will need knowledge of different statistical tools and what they can be used for. If you are not familiar with many different statistical tests, we’d recommend starting by learning what the following tests do:

• T-tests (paired and unpaired)
• Chi-Squared tests
• ANOVA (one-way, repeated measures and two-way)
• Simple Linear Regression
• Multiple Regression

You can learn more about these topics on our resources pages (link below)

Links & Resources to help

Introduction to Hypothesis testing from MASH

Resources about Hypothesis Tests Curated by MASH

MASH run workshops on hypothesis testing and getting to know the analyses listed above. Details are on .

---

Stage 4: Reporting the Results - communicate and display

What to do

This stage is all about summarising the results of your statistical tests, but adding context, discussions and conclusions.

Useful questions to think about answering when summarising your results are:

• What information answers your research question?
• Who/what is your audience?
• What statistical summarises do you want to present?
  • Descriptive summaries
  • Estimates, confidence intervals, p-values
  • Graphical summaries
• What discussions/conclusions can you draw?
• Are there any limitations in your research?
• What future research would you conduct?

At this point you want to put your results into the context of the research question and refer back to your null/alternative hypotheses. It’s important to report the accurate statistical findings, being transparent about the analysis you have conducted and any limitations of this. 

What are the statistical processes I need to know about?

Writing/reporting statistical research papers

Displaying data 

Links & Resources to help

Guidelines for Reporting Statistics -

Summarising Data Guide -

MASH run a workshop on data display. Details are on .