Experiment Design

Experiment Design

Experiments are an important part of research development and can be used to get valuable findings about a certain phenomenon. They can be done either in the field or in a laboratory. An experiment basically means the manipulation independent variables in different ways and then observes the outcome of the other dependent variables. Experiments can be done with animals or human beings as the variables. There different types of experiments depending on how experiment can assign subjects to different conditions. In this case the experiment is about testing the effect of e-mail request with different set of information and the effect on the responses (Ryan,2007).This experiment can be conducted as a within the subject or between the subject as discussed below:
Between the subject experiment:
This is one of the most common types of experiment especially when dealing with scientific experiments. This type of experiment is specifically designed to avoid carry over effects in an experiment. In this regard the participants can be in the control group or the treatment group within the experiment but will not belong to both hence each new treatment will need a new group (Godwin,2009).
In this case the experiment has been designed such that each offer presented has different benefits as part of modification to ensure that the experiment to derive the most accurate results and also ensure that the difference between the groups does not influence the final outcome.
Our case ahs picked on an e-mail message that invites people for a survey hence the survey is the main subject in this experiment and the participation by all the interested parties will derive the results to the researchers. To further differentiate the participants the e-mails have different offers which provide an opportunity for the participants to choose from. Tests will be conducted on the three categories of responses to determine which e-mail has the highest response numbers and to make its as the basis for the survey. The main interest in this experiment will therefore to find out which e-mail has the highest number of participants.
This type of experiment will therefore derive accurate results and free from contamination by extraneous factors this is because each participant will only be subjected to single treatment in terms of their preferences. This avoids boredom and even participant establishment which may affect the results from subsequent tests. This type of experiment will however require a large number of participants to be successful since they are only tested once. This is the major drawback for the between the subject design (Russell & Purcel, 2009).
Within the subjects design:
This is another type of experiment and serves in the opposite way as between the subject design and in this case the subjects or the participants in an experiment process are subject to more than one treatment which denotes the level of independent variables in the experiment. In our first design the participants were only subjected to a single treatment within the three e-mails and hence no single participant could be gauged on more than one e-mail. Under this new design the participants will be allowed to participate as independent variables in more than one type of e-mail and hence it’s possible to get outcomes that link a single participant to the hyperlink e-mail and also the $10 charity e-mail. This experiment will therefore test the participants on the three e-mails at different levels and hence it could be started with taking the survey on the hyperlink e-mail and determine the response and then subject the same group to the $10 charity e-mail and finally to $1000 chance e-mail. The outcome will then be analyzed and then reported in terms of the responses patterns in different cases. One of the key strengths of this design is its ability to reduce the error variance due to personal differences. This type of experiment also increases the statistical power and hence the beta error in an experiment which quite desirable in deriving accurate results in an experiment process (Johnson & Christensen, 2010).
Advantages of between the subjects design:
Although both experiment designs have advantages and disadvantages the between the subject model is the most proffered due to the following reasons:
One of the core strengths of the between the subjects experiment is that it has the potential or reducing the chances of contamination in an experiement.This is due to the fact the subjects are only allowed to participate in a single treatment with subsequent tests requiring new participants.
This design also allows a whole range of questions that will lead to accurate results. This is not possible with within the subject design because the same participants are used and hence its impossible to provide for a whole range of questions for treatment (Ray, 2008).
With these key advantages between the subjects experiment is the most common and highly valued by many researchers in social science and can lead to more accurate results.

References
Godwin,C.J.(2009). Research in Psychology: Methods and Design.6th ed. Hoboken, NJ : Wiley.
Johnson,B.& Christensen,L.(2010). Educational Research: Quantitative, Qualitative, and
Mixed Approaches.4th ed. Thousand Oaks, Calif. : SAGE Publications.
Ray,J.W.(2008). Methods toward a Science of Behavior and Experience.9th ed. Australia
; Belmont, CA: Thomson/Wadsworth.
Russel,B.& Purcell,J.(2009). Online research essentials: Implementing and designing
Research studies. San Francisco: Jossey-Bass.
Ryan,P.T.(2007). Modern Experimental Design. Hoboken, NJ: Wiley.

 

Leave a Reply