Quasi-Experimental Design: Estimating Intervention Consequences

Discuss about the Research Methods in Education for Collaborative Method.
 

 

The concept of quasi-experimental design is very popular with positivist researchers when conducting experiments in the research. Typically, quasi-experiment can be defined as an empirical study that can be applied by the positivist researcher to estimate the consequence of an intervention on the population under consideration. This design would be preferred by a positivist researcher because it estimates the impact of an intervention in the population under study without random assignment.

Unlike a true design, the quasi-experimental design allows the positivist researcher to monitor and control the assignment to the treatment condition. It also enables the researcher to adjust to an eligibility cut off the mark.  In the research to determine the effectiveness of Computer-Assisted Method, Collaborative Method and Lecture Method as methods of teaching mathematics in secondary school, the researcher preferred quasi-experimental design because it was simpler to set up as compared to a true design (Campbell, 2009). A true experimental design requires random assignment while quasi-experimental design lacks random assignment.

In addition, the researcher was trying to minimize risks associated with ecological validity since the quasi-experimental design is a natural oriented design, unlike a true experimental design that requires well-controlled laboratory settings. Again, the concept of natural setting in quasi-experimental design allowed the researcher to conduct a single experiment in the population and then use generalization rules to spread the impacts to other subjects and setting in the target population (Thyer, 2012). Moreover, this experimental design is more effective in longitudinal research thereby allowing the researcher to use long time periods to researcher different environments. Lastly, this experimental design minimizes the chance of conditional and ethical considerations that may affect the outcome of the experiment.

A null hypothesis refers to an assertion, a theory or a proposition that has not been proved. From a statistical point of view, it is used by the positivist researcher to indicate that there is no significant difference between the population under consideration and experimental errors (Haidt, 2006). It is denoted as H0. From the study under consideration, there are two null hypotheses that can be formulated to determine the effectiveness of Computer-Assisted Method, Collaborative Method and Lecture Method as methods of teaching mathematics in secondary school.

H0: There was no significant difference between the effectiveness of the computer-assisted method, collaborative method and lecture method in teaching mathematics in secondary school. If we assume the mean of the computer-assisted method is m1, mean of a collaborative method to be m2 and mean of lecture method to be m3, then our null hypothesis can be represented as; H0: m1=m2=m3.

In this case, a hypothesis can be formulated using the concept of correlation as formulated below.

H0: There is a correlation between the effectiveness of the computer-assisted method, collaborative method and lecture method in teaching mathematics in secondary school. If we assume the correlation coefficient of the computer-assisted method is δ1, the correlation coefficient of a collaborative method to be δ2 and correlation coefficient of lecture method to be δ3, then the null hypothesis can be formulated as; H0: δ1= δ2= δ3.

Internal validity can be defined as the approximate truth relating to inferences on causal relationships in the study population. This occurs when the researcher attempts to control dependent and independent variables that could affect inferences in the experiment. This exposes several threats to the internal validity of quasi-experiment. These threats include statistical regression, the participants, history, experimental mortality, selection, maturation, and testing. First, we consider history as a threat. This takes place when external factors to the subjects take place due to the passage of time (Maslow, 2013). Second, participants acts a threat of internal validity of the experiment because they drop out experiments before they finish (Marylene, 2014). Third, experimental mortality in internal validity of data occur due to geographical move and as a result of a different number of dropouts in the experiment. Fourth, testing is a threat to internal validity due to the effect of experience with protest thereby becoming test wise. Therefore, repeated testing will result to biases in the experiment thereby affecting the internal validity of the experiment.

The researcher would apply two techniques to ensure computer-assisted method, collaborative method and lecture method in teaching mathematics in secondary school have equal ability to deliver teaching objectives. They include matching technique, holding one or more variables constant, including an extraneous variable in the research design and analysis of covariance (ANCOVA). For the current research, we can use matching technique and analysis of covariance. First, in matching technique, the researcher can identify factors to be considered in the matching process. For this case, the researcher can select academic performance as a result of using the three methods of teaching mathematics (Kanungo & Manuel, 2014). The researcher can choose two top grades and two low grades from each group. Then to continue selecting until there is a clear match thereby enhancing an equivalent ability. Lastly, the researcher can apply the analysis of covariance technique. In this case, the researcher can adjust the dependents in the experiment and then equate the subjects to control and experimental group thereby enhancing equivalent ability.   

The process of accepting or rejecting null hypothesis must be determined by statistical test concerning the subject. Therefore, in order to reject the null hypothesis, statistical data analysis, and statistical tests must be carried out. As defined earlier in the context, the null hypothesis is a proposition that needs to be proved. To prove a null hypothesis, the researcher needs to compute various statistical tests so as compare with tabulated values (Schrage, 2014). The researcher may compute mean, standard deviation or covariance of each group. Then these values can be compared with tabulated values in various degrees of freedom in the statistical tables. Tabulated values may be obtained from chi-square tables, ANOVA tables, t-test tables or F-test tables. Therefore, if the calculated value is less than tabulated value, the researcher is allowed to reject the null hypothesis. In the current research, the null hypothesis will be rejected if;

Calculated value

An attitude scale us used to provide accurate and valid social attitude towards a particular issue. It is designed to measure the attitude of every individual concerning a particular matter in the public domain. There are five items in attitude scale that can be applied towards recycling of wastes. The first item that can be designed is ‘my friend_ recycle’. In this item, the main perception is based on individual attitudes towards recycling of wastes (Arnold, 2010). This individual is using personal based perception to design an item that their peers can join thereby creating a perception of the social norm in the recycling of wastes. This demonstrates a perfect positive correlation between attitudes and recycling of wastes at an individual level. For this item to be effective, it is very important to measure and explore the relationship between attitudes and behavior of an individual to recycle or not to recycle wastes.

The second item to be considered in attitude scale is ‘I want to recycle more than I do now’. This item is also very applicable to measure the attitude towards recycling of wastes. This item focuses on the motivational level of an individual towards recycling of wastes. This item tries to determine the Importance of recycling of wastes to an individual. If the individual is propelled towards recycling of wastes by his or her attitude, it is statistically correct that there is a perfect positive correlation between the attitude and the importance of recycling by an individual.

The third item to be designed is the perception of social norms towards recycling of wastes. It also relates to the subjective norm in towards the recycling of wastes (Thomas, 2009). In that connection, if the attitude towards the recycling of wastes is driven by social and subjective norms, we conclude there is a positive correlation between recycling of wastes and social norms.

The fourth item that can be designed to measure the attitudes towards recycling of wastes is perceived behavior control. In this item, the individual is able to control the behavior so as to determine the attitude towards recycling of wastes. In that case, the correlation between recycling of wastes and perceived behavior control is high (Richard, 2013). This item develops a specific attitude that individuals and their peers can generate towards recycling of wastes.

The last item to be considered in this case is behavior intention. In this case, the researcher may try to determine the relationship between recycling attitudes and behaviors (McGregor, 2012). This will be used to measure how effective people would use intrinsic motivation to improve their attitudes and behaviors in the recycling of wastes. The researcher would examine the environment and the concept of attitudes towards recycling of wastes so as to determine whether to accept the hypothesis or not (Ryan & Deci, 2017). In conclusion, if all the five items indicate a good measure of attitude towards the recycling of wastes, the null hypothesis would be accepted. That is, there is a relationship between attitudes and recycling of wastes.

In order to ensure that the attitude scale measures the attitude towards the recycling of wastes, it is important to consider the effectiveness of attitude scale. First, it is very important to collect relevant data on specific attitudes, behavior intention, subjective norms and perceived behavior control (Singh, 2015). This will facilitate the research to formulate and test the hypotheses. The method of collecting data should be free from biases to minimize data manipulation and inconsistency of data. This will ensure that the attitude scale measures the required attitudes towards recycling of data. Again, the effectiveness of attitude scale is improved by analyzing the collected data through the use of statistical methods such as Pearson chi-square test, ANOVA, and statistical distributions. If the correct data analysis methods are applied, then the interpretation of data will be appropriate. This will ensure data is presented as analyzed thereby ensuring attitude scales measure what they are intended to measure. Lastly, it is very important to consider semantic deferential technique in measuring attitude so as to determine potency, evaluation and activities involved in attitude scale (Edwards, 2009). Through evaluation, the scale will determine the positive and negative side of a person towards the attitude on a given topic.  Through potency, the topic will be measured in terms of strengths and weaknesses. Lastly, through activity, attitude scale will measure the passive and active part of the topic. These factors will collectively enable attitude scale to measure what it is intend to measure.

Arnold, J. (2010). Coaching Skills for Leaders in the Workplace: How to Develop, Motivate and Get the Best from Your Staff. How to Books.

Campbell, D.T. (2009). Experimental and Quasi-Experimental Designs for Research. Cengage Learning.

Edwards, A.L. (2009). Techniques of Attitude Scale Construction. Irvington Pub.

Haidt, J. (2006). The Happiness Hypothesis: Finding Modern Truth in Ancient Wisdom. Basic Books.

Kanungo, R.N., & Manuel, M.  (2014). Work Motivation: Models for Developing Countries. Sage Publication put.

Marylene, G. (2014). The Oxford Handbook of Work Engagement, Motivation and Self-Determination Theory. OUP USA.

Maslow, A.H. (2013). A Theory of Human Motivation. Start Publishing LLC.

McGregor, D. (2012). The Human Side of Enterprise. New York, 21.

Richard, A. (2013). Job Satisfaction from Herzberg’s Two Factor Theory Perspective. Grin publishing.

Ryan, R.M., & Deci, EL. (2017). Self-Determination Theory: Basic Psychological Need in Motivation, development, and Wellness. The Guilford Press.

Schrage, M. (2014). The Innovator’s Hypothesis: How Cheap Experiments are worth more than Good Ideas (MIT Press). The MIT Press.

Singh, P. (2015). Construction of Attitude Scale. LAP LAMBERT Academic Publishing.

Thomas, K.W. (2009). Intrinsic Motivation: What Really Drives Employees Engagement. Berret-Koehler publishers.

Thyer, B. (2012). Quasi-Experimental Research Designs (Pocket Guide to Social Work Research Methods). Oxford University Press.