Advanced Design in Nursing Research, Essay Example

Subjects: The first population, 50 couples that were administered training on money management techniques at a workshop whereas the second population was not provided training.

Design 1- This study used a quasi-experimental Nonequivalent Control Group Design with two groups: (1) Experimental group A (First population administered training on money management techniques at a workshop- Treatment Group), (2) Experimental group B (Second population was not provided training- Control Group)

Characteristics of the design: In this case we evaluate with the Static Group design where two intact groups are tested after one has received training. Now none of them have been equated prior to treatment.  Thus the populations being compared are not equivalent on all things prior to the treatment. This heightens risk of internal validity. When a post-treatment difference between groups is observed, one cannot be confident enough about the treatment, since the groups had pre-existing differences before the observed post-treatment difference.

Procedure

1. Administer Post-test to both groups.
2. If the result favors the treatment group- i.e. the couples who received the training on money management techniques have less arguments and disagreements about money. It can be interpreted (though less so than with random assignment) it was due to the program.
3. If the result shows no variation then it implies that program is not effective.

The dependent variable for this design is the disagreement about money experienced between the couples. The controls for the design are couples’ age, education, number of children, and length of marriage, wives’ employment and other characteristics. The null hypothesis will be rejected.

The design for this study will simply be plotting the results on a chart:

Figure 1

Design 2

Subjects: Randomly selected 300 married couples from marriage licenses in 10 states.

Design: This design is a non-experimental design because is not manipulating the independent variable of the couple’s money management knowledge. The survey uses a questionnaire to obtain results to study. The control variables are the couples’ ages, length of marriage, education, number of children and wives’ employment. The dependent variable is the disagreements about money for the couples. This study does not make a hypothesis of the results of the study and dependent variable. It is simply asking questions about money management.

Method: The design will use the chi square test to test the validity of the study results.

The number of responses from the couples with better management, 4, is compared to the number of possible answers for better management of money, 8, and the number of expected answers to reduce arguments, 4, to the number of possible answers for argument results. This chi square value would be compared to the criterion value for the alpha of p<.05 and the degrees of freedom.

The study results in a symmetric curve that shows the equal amount of results for the 300 married couples. The results are displayed here in a bar graph:

Figure 2

Design 3

Subjects:  Study recruited 200 married couples from a University population of married student couples (via flyers, ads in the student newspaper, etc.) and randomly assigns these couples to groups.

Design:  Here we use the Solomon Four-Group Design which helps to deal with a potential testing threat. Note that the design has four groups, two of the groups receive the treatment and two do not. Further, two of the groups receive a pretest and two do not. In our case study, 200 married couples from a University population are taken in a process of random assignment. Some of these couples were given training and others were not. Some of the couples were just presented in a controlled area and observed for responses to money management and disagreements. The independent variable for this design is the training that is given to some of the couples and not to others. The independent variable consisted of the videos, unsolved problem one and unsolved problem two. The dependent variable is the couples’ extent of disagreement over money management. The control variables are the room and other controlled conditions the University personnel employed.

Method: The model is an integration of the pretest-posttest control group design and the posttest only control group design.  It is very useful to control threats to internal and external validity just like the only posttest control group design. It provides the strongest design to evaluate the effectiveness of the intervention and allows the assessment of confounding variables and testing effects.

Figure 3

The design will be statistically tested using a parametric statistical test. The value assigned to the results for each of the four coupled groups (t-value) will be compared with all the different possible results these couples responded with. This will define the degrees of freedom. The criterion value for this study was +- 1.96 Standard Deviations in summarized responses. The results of this study revealed that the t-value was less than 5% of the time by chance alone that the couple’ money management improved. Therefore, an alternative hypothesis was accepted that when couples are trained in money management techniques, the disagreements about money are reduced.

Non Experimental Model

• Advantage  : Number of human characteristics/ independent variables are not subject to experimental manipulation or randomization
• Advantage : Some variables cannot ethically be manipulated
• Advantage : Non experimental research is often needed to scope out the experimental one.
• Disadvantages: Information seems to be superficial as breadth is emphasized.
• Disadvantages: Costly and time consuming

Solomon Group model

Advantages: it can be used in combination of the pretest control group and the posttest only design. I would prefer this model since it combines the two designs with the addition of able to test and control for instrument reactivity. Moreover it helps to assess the implication of pretest sensitization. Thus it helps the researcher to have more confidence by reducing the internal and external threats to the design process (Brink & Wood, 1989)

Read the study: (20 points)

Borden, L. M., Lee, S., Serido, J., & Collins, D. (2008). Changing college students’ financial knowledge, attitudes, and behavior through seminar participation. Journal of            Family and Economic Issues, 29(1), 23-40.

1. Critique the research design they used, citing both positive and negative aspects. Assume the researchers wished to draw causal conclusions.
2. Plan and describe another research design for answering the same research questions they addressed. Attempt to improve any aspects of the research that you critiqued.

Type of Research: Descriptive Pre Experimental Research

Research Design: Correlational       

Pretest –Experiment- Post test (Absence of any control group)

This a Pre Experimental Correlational Research model, since the study is made without manipulation of treatments or random assignment. A Correlational research design is used to describe the statistical association between two or more variables. It involves collecting data in order to determine whether, and to what degree, a relationship exists between two or more quantifiable variables. They observe participants in order to describe them as they naturally exist without experimental treatments. Non experimental research is used when variables of interest cannot be manipulated because they are naturally existing attributes or when random assignment of individuals to a given treatment condition would be unethical. Here the variables of the study are not manipulated but studied. Thus the researcher here needs to consider possible alternative explanations, to jointly analyze the several variables, and to present conclusions without any definitive casual statements.

Positive aspects

• A research design in which a researcher observes a phenomenon without manipulating the independent variables(s).
• No manipulation.
• Independent variables have already occurred, so no control over them.
• Clear, concise problem statement that is based on a theoretical framework, or natural phenomenon.
• It is appropriate to describe or predict behavior.

Negative aspects

• In these studies is that it is impossible to determine a cause and effect
• Random assignments of samples are not possible.

Suggestion

Since the main flaw is the absence of random assignment and control group, I would suggest a model which has both these parameters, like the Solomon experiment model. The pre experimental model is flawed with internal validity- while if we use the Solomon model it is comprehensive and manages the internal validity and the external validity well. So we get a better cause effect relationship and the research is more pragmatic.

References

Bordens, K., Abbott, B.B. (2008). Research Design and Methods: A Process Approach, 7^th edition. New York: McGraw Hill Higher Education.

VassarStats. (n.d.) T-test for independent samples (n per sample up to null). http://faculty.vassar.edu/lowry/t_ind_stats.html

Brink, P. J. & Wood, M. J. (1989). Advanced design in nursing research. Newbury Park: Sage Publications.