Experimental and quasi-experimental design - Term Paper Example

? Experimental and Quasi-Experimental Design Advantages and Disadvantages of Experimental Design There are various advantages of experimental design, with the first advantage being the creation of a cause-effect relationship. This design gives researchers or investigators sufficient control to establish and analyze an effective relationship. By setting up experiments and executing them, investigators can find the real reason for different occurrences (Frankfort & Nachmias, 2008). This enhances the understanding of reasons why events happen, and helps predict any future behaviors. It provides insights that may be used to bring results in respect of an immediate concern. For example, by experimenting with different teaching methods, a teacher can come up with effective teaching plans and use these results to establish the most appropriate option to use and increase benefit to students. It provides reliable results by utilization of procedures, quantitative measurement levels and random selection. This assures the researcher that the outcomes obtained are a true representation of the real occurrence. The first disadvantage of experimental design is that it involves abuse of ethical standards. In order to get the expected results, the researcher may manipulate people’s revelations to the study being conducted. For example, people believe children’s exposure to violence on television increases aggressiveness, thus the researcher may derive justification to deliberately expose children to such violence. Excellent experiments are hard to conduct since they require a lot of resources and human energy, which may not be readily available. A major disadvantage is that an experiment usually takes the behavior the researcher is interested in out of context (Creswell 2009). This may produce substantial artificiality and questions on how readily the findings can be generalized in other contexts. Lastly, it takes cleverness, ingenuity and experience to design an experiment well and some characteristics may be lacking, especially specialization. 2. Advantages and Disadvantages of Quasi-Experimental Design The first advantage of quasi-experimental design is easy conducting of logistics. The researcher does not require to group individuals as they are pre-grouped. For example, different classes can be selected and the treatment can be given to the experimental groups, which are usually predetermined in class or school, thus easing the researcher’s study. This offers a possibility of comparison between the experimental and control group (Creswell, 2009). The control group does not receive any treatment while the experimental group gets exposure to a treatment, and afterwards, the two groups’ performances can be compared after data collection to determine any differences in their performances after treatment. This is a key advantage since it helps the investigator to make inferences on the possible existence of a reason and determine a relationship between the treatment and outcome. There are various disadvantages of quasi-experimental designs, the first one being the possibility for non-equivalent groups. The heterogeneity in the research group may be undesirable, since using a sampling method other than random sampling will increase the potential for constructing groups that are non-equivalent. This is due to researchers obtaining control and experimental groups that are similar by random selection. The quasi-experimental design has a low potential for internal validity since the effects and causes of the phenomenon being tested can be obvious (Frankfort & Nachmias, 2008). Starting research with non-equivalent groups poses a threat to internal validity of results (Creswell, 2009). By not starting with equivalent groups, the researcher cannot offer assured results that the treatment involved is the factor causing the change since other factors may have brought in the change. 3. Describe the Pros and Cons of Cross-Sectional and Quasi-Experimental Design Relating to Their (Internal vs. External Validity) a. The pros of Cross-Sectional Design in Relation to Their Internal and External Validity Internally, cross-sectional design is efficient in evaluating large samples. This helps in understanding the prevalence of specific services and generates hypotheses of service delivery across different sections. In terms of design, it estimates the association between provision and ownership of specific types of services (Creswell, 2009). For example, characteristics are portrayed by a caregiver and patient according to social demographics and gender or age respectively. Externally, it offers a possibility of comparison between the control and experimental group. This offers accuracy in results and is fast and effective. b. The Cons of Cross-Sectional Design in Relation to Their Internal and External Validity Internally, cross-sectional design is weak in terms of history; this is behavior measurement at different points, which reflects the impact of independent variables (David, 2001). Cross-sectional design has no control over these occurrences as a result of change in time and cultures. The longer the time elapsed between measurements, the riskier history effects. In terms of maturation, it produces different changes over a period of time, such as growth in the nervous system and fatigue which produces behavioral changes to experimental variables. Externally, it is weak in terms of intervention and interaction of the selection. Cross-sectional design bias in sample selection poses a threat because of the time and cost committed, thus being devastating to conclusions drawn in the study. Comparing individuals of different age groups on the same opinion or task offers a risk of behavioral differences resulting from different generation views, education and culture. c. The Pros of Quasi-Experimental Design in Relation to Their Internal and External Validity Quasi-experiments are very close approximations of accurate experimental designs and due to this, they can be used to investigate relationships among factors. In relation to validity, the quasi-experimental design has a low potential for internal validity since its effects and causes can be obvious, and in research these can be subtle. Externally, they are the most effective when random assignments are not practical or do not make sense in relation to the research population or ethical considerations. It has validity that helps the investigator make inferences on the possible existence of a reason and achieve relationship of the treatment. d. The Cons of Quasi-Experimental Designs in Relation to Their Internal and External Validity Testing: This is a threat to reactivity as a result of learning or practicing from repeated testing exposure. The persons taking the tests are subject to this threat over a period of time. Statistical regression posses an internal threat to research as it occurs where participants are selected, since they are extreme on classification of factors of interest such as intelligence. The results tend to drift to the average rather than become extreme, thus regression effects will obscure development changes and obscure treatment. Quasi-experimental designs face instrumentation problems whereby researchers let testing instruments and studies become invalid because of change in time or culture. These changes introduce different observers or techniques leading to altering of the continuity of measurement. Externally, quasi-experimental designs face weak validity in terms of interaction of intervention and testing (Rossi & Freeman, 2003). This requires participants to get involved in a number of tests on different occasions and at different ages. This will pose a problem of different opinions due to different settings and the age factor. References Creswell, J. W. (2009). Research design: Qualitative, quantitative, and mixed methods approaches (3rd ed.). Thousand Oaks, CA: Sage Publications. Frankfort-Nachmias, C., & Nachmias, D. (2008). Research methods in the social sciences (7th ed.). New York: Worth Publishers. Rossi, P. H., & Freeman, H. E. (2003). A systematic approach. Thousand Oaks, California: Sage Publications. Vaus, D. de (2001). Research design in social research. Thousand Oaks, California: Sage Publication. Read More