Conducting a Study and Statistics Planning Study Guide

Updated on Oct 5, 2011

Introduction to Conducting a Study and Statistics Planning

Statistics involves the collection and analysis of data. Both tasks are critical. If data are not collected in a sensible manner, no amount of sophisticated analysis will compensate. Similarly, improper analyses can result in improper conclusions from even the best data. A key to a successful study is to establish a solid framework. In this lesson, we will outline such a framework and discuss the types of inference that can be made from different types of studies.

Steps in Planning and Conducting Studies

Most studies are undertaken to answer one or more questions about our world. Would drilling for oil and gas in the Arctic National Wildlife Refuge negatively affect the environment? Do laws mandating seat belt use increase the rates of their use? Is the flu vaccine safe and effective in preventing illness? These are the types of questions statisticians like to answer.

Planning and conducting a study can be outlined in five steps, each of which we will discuss briefly:

  1. developing the research question
  2. deciding what to measure and how to measure it
  3. collecting the data
  4. analyzing the data
  5. answering the question

Developing the Research Question

Statisticians often work in teams with other researchers. The team works together to determine the research question to be addressed in an upcoming study. To fully specify the research question, the study population should be identified and the goals of the study should be outlined. The statistician must understand the question(s) and the goals of the study if he or she is to be a full member of this team.

Deciding What to Measure and How to Measure It

Once the research question has been specified, the team must determine what information is needed to answer the research question. Identifying what variables will be measured and deciding how they will be measured is fundamental. Sometimes, this step is obvious (as in a study relating salaries of individuals to educational level). At other times, this is extremely challenging (as in a study relating attitudes toward school to intelligence).

In some studies, a comparison of two or more regimens or procedures may be the focus of the research question. As an illustration, a study could be used to determine whether students perform better on English tests if they study in a quiet environment or while listening to classical music. To answer the question, some students would study in a quiet environment; others would study while listening to classical music. The scores on the English test for each group would be used to answer the research question. The study environments (quiet or classical music) would be the treatments in this study. A treatment is a specific regimen or procedure assigned to the participants of the study.

Collecting the Data

Good data collection is a crucial component of any study. Because resources are always limited, the first question is whether an existing data source exists that could be used to answer the research question. If existing data are found, the manner in which the data were collected and the purpose for which they were collected must be carefully considered, so that any resulting limitations they would impose on the proposed study can be evaluated and judged to be acceptable. If no existing data are found, a careful plan for data collection must be prepared. The manner in which data are collected determines the appropriate statistical analyses to be conducted and the conclusions that can be drawn.

View Full Article
Add your own comment