Experimental Design (15%) - Purdue University



Experimental Design (15%)

Topic and title must be approved by me no later than Monday November 10th. The project is due December 8th by 5 p.m. Late projects will be docked 10% each day they are not turned in.

Purpose: The purpose of this project is to get you to think about research, and to learn about experimental methodology. Our field is relatively new, and there are so many questions that we still have no or incomplete answers for.

Experimental Scientific Method

Experiments help us investigate “cause and effect” relationships by allowing us to systematically test a particular relationship. A scientist (or student) should initially choose a specific question or problem to study that can be formulated in terms of a hypothesis (educated guess) that can actually be tested. In scientific lingo, tests done to check whether or not a particular hypothesis holds are called experiments. Variables are things that will affect the system you want to investigate. Independent variables (or manipulated variables) are the things that will be the CAUSE of the changes you measure. A dependent variable is the variable that responds, or the variable you will measure after the experiment is set up. It will be the EFFECT of the action you took. Figuring out what your variables will be requires thought, information gathering, and a study of the facts related to your problem. The following steps will help you systematically investigate a question/problem that can be tested.

1. Idea Formulation

Formulate a specific question(s) or problem(s) that will be the basis for your experiment. Are you interested in a particular disordered population? How different therapies improve speech production? The way one of the speech subsystems (respiratory, laryngeal, articulatory) functions during normal/disordered speech?

2. Information Gathering or Literature Review

This is where you will find out what others’ have already learned about the particular area/question you are interested in. Search for books at the library related to your topic, do a PUB MED search entering specific and relevant terms (i.e. Parkinson’s Speech Therapy Techniques). Initially browse through abstracts to see if a particular study is important and then go get the full article.

3. Title the Project

Choose a title that describes your investigation. The title should be concise but should communicate what the experiment is about. If you are stuck you can always fall back on:

“The Effect of (Independent Variable) on (Dependent Variable)”

4. Statement of Purpose a.k.a. Goal Statement

What do you want to find out? Why is your experiment important? Use your ideas/questions to formulate a clear sentence about the why/how behind your experiment. It will be similar to your title, just perhaps a bit more in depth. An example:

“Because there is conflicting/little information about the effectiveness of Therapy A and Therapy B, this study was designed to determine how these different therapy approaches improve overall speech intelligibility in individuals with Parkinsons Disease (IPD)”

5. Identify Variables

Based on your gathered information, decide on or make an educated guess about what types of things affect the system or area you are investigating.

What are your independent variables or “causes”? When multiple variables are used, they may somehow be linked or work together to cause something. Therefore it may be difficult to parse out which independent variable is specifically responsible for whatever change you measure with your dependent variable. Therefore, you need to think about what controls you will use, or how you will ensure that your independent variable(s) is responsible for changes to your dependent variable.

What will you measure, or what is the dependent variable(s)? I will put a few books and perhaps an article on reserve that provides an introduction to instrumentation used in the field of speech/hearing sciences. This is just to familiarize you with some dependent variables, but by no means should you feel limited to use these. Remember, a dependent variable can be something as simple as a measure of intelligibility (i.e. the percentage of words spoken by a IPD that are correctly identified by a naïve listener).

6. Make Hypothesis

When you figure out what variables will be involved in your experiment, and how you will be able to identify which variable(s) is causing your observation, you should now work on translating your question(s) into a hypothesis. A hypothesis is a question that has been reworded into a form that can be tested by your experiment. Use what you’ve learned from reading books, articles, etc. as well as personal intuition to make an educated guess about for a cause and effect relationship or what the outcome of your study may be. Only by actually running an experiment and analyzing the data can a hypothesis either be accepted or rejected. I just want you to gain experience coming up with a hypothesis (or hypotheses).

7. Design Experiment to Test Your Hypothesis

Make a step-by-step list of what you will do to test your hypothesis. This part is really the ‘Methods’ section or the “How” of a research article.

Be sure to include information about:

Participants (subjects) in the study: Who will be in your experiment? How many participants will you use? Will you have males and females? What ages will you include? Who will your control subjects be? How will your subjects be selected? How will you screen the normal or disordered participants to ensure that they really represent the population they are supposed to represent?

Protocol of the study: What special things/equipment will you need for your experiment? What will you have you actually have your participants do? What will be your specific tasks and/or procedures? How will you collect your data? How many times will the experiment be run to ensure valid results and thus, control for naturally occurring experimental error?

Independent Variables (Causes) and Dependent Variable(s) (Effect—what you will measure)

Specific Requirements for the Project

1. Scientific Journal

This can be a handwritten or typed informal journal detailing how you came up with your idea to how you decided your experiment should be run. By reading your journals I should be able to see exactly step-by-step timeline of how you started and finished the project. Each time you meet with your group members someone should scribble down the “minutes” of your meetings. If you work on something alone, you should scribble down the process so I can a clear idea about what everyone’s role was in the project. The journal can be a binder with loose-leaf pages, a diary, a spiral notebook, etc.

2. Experimental Paper

The actual paper should be typed, double-spaced, in 12-point font, and should be no less than 6 pages (including a bibliography page). It should include the following sections:

I. Introduction

• Literature review. Give me a brief history of what other researchers have found, or what is known thus far about your topic. This write up should include no less than 4 sources of information from books, journal articles, etc.

• Statement of purpose (see above)

• Hypothesis (see above)

II. Methods

• Participants (see above)

• Protocol (see above)

• Variables (see above)

III. Bibliography or Reference List

• Alphabetical listing of the articles you cited as well as anything else you used (reserve materials) to complete the project. Refs should be in APA format!!

................
................

In order to avoid copyright disputes, this page is only a partial summary.

Google Online Preview   Download