Design and Implementation - Assignment Example

?Design and Implementation Contents Question 3 Question 2 4 Question 3 5 Question 4 6 Question 5 6 Question 6 7 Question 7 8 Reference 10 Question 1 What does “randomization” mean in the context of a conversation about randomized experimental designs and what are three reasons for using “random assignment?” Answer An experimental design can be referred to as a means of attaining the results of a psychological research. A randomized experimental design includes an independent variable with more than two levels of experimental conditions. Here each individual has the equal probability of being selected at every draw. The process of indiscriminately choosing the groups or individual is known as randomization. The objective of such action is to reduce bias within the study. Randomization is mostly used in the medical field where testing the effects of a new medicine is done through selecting equal number of people for the experiment. These people are classified as control and experimental groups. The control groups are not exposed to the medication while the experimental group receives some amount of medication. Random assignment are widely used throughout various surveys because- 1. Randomly assigning the participants for the experimental study prevents the researcher from subconsciously influencing the sampling units for providing them a particular type of treatment. In other words, randomization ensures a liberal and reliable study where the results will not vary even if the researcher gets replaced somehow. 2. A random experiment equally distributes the nature of the groups, participants or subjects which are being studied. This equal distribution prevents the outcome of the experiment from being biased. However the chances of coincidence happening in case of two people belonging to a particular group affected by similar uncontrolled conditions can give rise to a biased outcome even if the randomized method is employed. 3. Random assignments allow the application of statistical tools which needs particular characteristics related to the information. This method presents a system to obtain probabilistic characteristic of the data by the efficient elimination of the effects of the irrelevant and unnecessary factors. Question 2 Describe the characteristics of a between-subjects (or, nested) design in your own words. Provide an example of how you could use a nested design to conduct a study on a topic of interest to you. Answer Between-subjects design refers to such experimental designs where the participants are exposed to any one condition from among the various conditions. However in this case, a participant cannot be allowed to go for both the conditions. In other words, between-subjects design is used for measuring the effect of a particular thing on different groups, exposing one group to the particular condition and restricting the other group from being exposed in such condition. It never exposes the same subject to different conditions for studying the changes in the outcome. The statistical term for this method is nested design. This method can be applied in the process of measuring the effectiveness of the teaching aids used in modern times. In this experiment, the participants will be grouped randomly independent of their age, gender or educational status. Two groups will be formed namely group A and group B. Group A will go through a learning session where traditional methods like books, lectures and weekly tests will be included. The other group (Group B) will be exposed to audio-visual mediums, internet and group-discussion. After a month, both these groups would seat for an exam and the results will decide which teaching method is more effective. But it should always be noted, that factors like intelligence quotient (IQ), retention capability, slow learning, age etc. plays a major role in the outcome of the test. Question 3 In your own words, describe the characteristics of a within-subjects (or, repeated measures; or, crossed) design. Provide an example of how you could us a repeated measures design to conduct a study on a topic of interest to you. Answer A within-subjects design is a kind of experimental devise in which every participant is exposed to every condition or treatment. The term treatment in this case, implies the various levels of the independent variable. This design doesn’t require a wide pool of participants for conducting the study and the chances of errors are comparatively lesser than the between-subjects design. However there are chances of difference in the attitude of the participants due to the effect of the previous conditions to which they had been exposed and this particular flaw can lead to defective outcomes. This design can be applied to measure the changes taking place within a group of people before and after the consumption of an energy drink. The group would be exposed to certain tasks that require both physical and mental strength. The tasks will come in a series and the group would try to complete those within time. Now we can measure the time taken to complete a task before having the drink and the time taken to complete the same after having the drink. The measurements would be compared then, to find out whether energy drinks actually affect the energy level of individuals or not. Question 4 What is a factorial design and what is the key strength or advantage of this research design type that is not available in a single-factor design? Answer A factorial design is the means to study the significance of two or more independent factors on the response group. In a factorial design, every level of each independent variable is combined with every level of the other independent variable to create every possible condition. The basic benefit that can be derived from this design is that a two-way method helps us to observe the joint impact of the independent variables on the dependent variable. A joint impact implies that the effect of one independent variable on a dependent variable is not equal for every level of the other independent variable. This information can’t be obtained by the application of single-factor design. Question 5 What is the greatest potential problem in a within-subjects design with repeated treatments and repeated measurements? What do Rosnow and Rosenthal (2008) recommend as an effective strategy to address the problem? Answer The major problem that often takes place in a within-subjects design is that the arrangement according to which the participants are exposed to the various treatments are often confused with the effects of the treatment. For example, when a group of children consume some health drink (as part of the experiment) they automatically exhibit a tensed feeling for the first time and then they would get accustomed with the upcoming phases. However their tension and nervousness would play a vital role in the tasks performed by them. For addressing this issue, the method of counterbalancing has been introduced. Counterbalancing refers to rotating the sequence of the treatments. This means a few children will consume health drink before performing the tasks assigned to them and then they would be offered with fruits and vegetables later on. For other kids, the sequence will be reversed which means they will eat some fruits first and then they will be offered with the health drink. Latin square design is a famous statistical design with built-in counterbalancing tool. Question 6 Identify and describe three criteria for inferring causality. Answer The three main criteria for inferring causality are- 1. Co-variation- It is referred to the fusion of contiguity and constant conjunction. However it is also considered that the conjunction between the cause and effect is not constant always but it is likely to be constant (Rosenthal and Rosnow). 2. Temporal precedence- It refers to the priority and assumes that the effect of an event is always preceded by the cause. 3. Internal validity- It is the attempt of the scientists to eliminate the parallel explanations for the conjunction on the basis of empirical and logical grounds. Question 7 Look back at the article posted in Unit One, Lesson Two by Principe, Kanaya, Ceci and Singh (2006). Pay particular attention to the method section of the article. Is this study an example of any of the pre-experimental designs presented in Rosnow and Rosenthal (2008)? If so which one? What are some of the threats to internal validity in the study by Principe et al. (2006)? Explain how you identified the potential threats to internal validity and suggest ways that these threats might be addressed in a follow-up study. Answer It can be said that the methodology applied in the given article follows “One shot case study” method which is represented by X-O, where X refers to exposure to an experiment variable and O refers to the observation. A major threat to internal validity is the absence of the highly suggestive questioning technique which would have helped the children to recall their pasts. Also there are high chances among kids to imagine events which have been witnessed by their classmates and eventually thinking that they themselves had actually experienced those events. Internal validity confirms the reliability of the causal conclusion drawn from the study. In the given case the reliability highly depends on the information provided by the kids and therefore the internal validity is equally dependent on the same data. When the internal validity faces such threats then the researcher must try to eliminate the impact of the extraneous variables as far as possible. Also verifying whether there actually exists any cause and effect relationship between the event and the outcome might lead to accurate results. Reference Rosenthal Robert and Ralph Rosnow, Essentials of Behavioral Research: Methods and Data Analysis. New York: McGraw-Hill higher education, 2008. Print. Read More