Analysis of Non-Experimental Correlational Designs

2205 words (9 pages) Essay

18th May 2020 Data Analysis Reference this

Tags:

Disclaimer: This work has been submitted by a university student. This is not an example of the work produced by our Essay Writing Service. You can view samples of our professional work here.

Any opinions, findings, conclusions or recommendations expressed in this material are those of the authors and do not necessarily reflect the views of UKEssays.com.

Introduction

    Correlation is a measure of the relationship between variables. Correlational designsare used when researchers need to explore the relationships that may exist amongst variables that are not manipulated or cannot be manipulated (Fitzgerald, Rumrill & Schenker, 2004). An establishment of variables could man a relationship between many variables or could be as simply one variable having a relationship to another variable. Formation of a relationship could be either be a positive or negative one. In a positive relationship, as one variable increases, the other variable will increase also. In a negative relationship, one variable decreases as the other variable will increase. However, correlation is not a measure of interconnection but can be an indicator of the existence of a relationship between variables. Correlation can have the possibilities of a range from -1 to +1 with both limits indicating a component of a strong relationship. In there is an absolute value present above 0.5 it indicates a strong relationship while an absolute 0.5 indicates an average relationship between variables. When the values fall below 0.5 this will indicate the existence of a weak relationship between the variables.

Strengths of Correlation Designs

    When looking a look at how a study can determine the validity of a correlation design you have to be able to understand the strengths and weaknesses. A correlation can establish the existence or nonexistence of a relationship between two factors where it then can be used for representing areas where experimental research may take place and show further outcomes. Therefore, researchers can gather much more data the experiments.

    When a researcher chooses to use correlational design, they can gauge the extent of relationships between important variables.  Naturalistic observation is an advantage of correlational studies.  To be able to see what is happening in real-time it can be captured with a naturalistic observation by the researcher. Rather than having to study photographic images or playback of a video recording of an incident.  They will have an opportunity to observe the phenomenon as it transpires. As the action happens and change occurs the researcher is present in the moment to capture data that will be later analyzed to provide evidence to support their concern (Mertens, 2014). An example, the researcher could observe a group of middle school students interact with each other in a way that is seen from many angles and record what is seen in real-time. This will also allow the researcher to see the action and their behavior patterns in natural settings such as a classroom or outside in a play area. When conducting a deeper investigation, variables relationships can be examined.  The researcher may realize other additional information could be appropriated and used through observation that was not part of the thought process of gathering data originally. 

    When trying to find a relationship between Factors Related to Academic Success among Nursing Students, we find a good example of correlation analysis (Beauvais, Steward, DeNisco, and Beauvais, 2014). Looking at the analysis of relationships between variables in this case study is substantial given there is a test of significance of the correlation. Standard practice is to statistically determine whether the correlation between variables occurred by chance or truly there was sufficient evidence to support the claim for presence of correlation. Since it was an established test of significant the researchers were able to conclude for example that better resilience could have some influence on better academic performance.

Weaknesses of Correlation Designs

    While conducting correlational research it takes into an account the linearity in relationships.  If the established relationship is not linear, the strength of it can be diminished which can, in turn, alter the results of the research (Creswell,2014; Creswell & Creswell, 2018). If the amount of data that will be analyzed is decreased the overall strength of the research could be vulnerable.

 There could be numerous concerns with all research studies that are conducted. Likewise, when conducting Correlational research some studies will present concerns too.  One of the chief concerns within correlational studies that may appear is when researchers attempt to understand the data that has already been collected from other studies.  Multicollinearity could raise concerns with predictive studies.  Multicollinearity may have several adverse effects on estimated coefficients in a multiple regression analysis; consequently, it is important that researchers be trained in detecting its existence (Mansfield & Helms, 1982). The concern is amplified when there are copious predictor variables that are factored in the study. The possibility that more than one variable can be closely related to another multicollinearity occurs.  Once this occurs researchers cannot say without uncertainty which variable will impose the greatest influence. Since the variables are very comparable and may host some of the same features, the researcher may not be able to distinctly state which variable caused the change to ensue. The researcher will undoubtedly want to use methods that would manufacture the maximum amount of accurateness.  Without having accurate data, the researched information can be invalid, unreliable or unnecessary manipulated (Fitzgerald, Rumrill & Schenker, 2004).  If fewer predictors are used the research may produce better outcomes. 

     An account of another weakness of correlational designs comes in the form of quality and consistency of data collection.  When collecting data that is going to be used for research purposes it should be gathered in the most relevant manner that will yield the wanted results.  The quality factor should always take precedence when conducting research studies.  Just having a few quality samples could be enough rather than having a high quantity of samples with very diminutive quality.  If the record-keeping is not conducted accurately it could pose another problem for the correlation designs. The note-taking that will be in ongoing can be easily recorded incorrectly and therefore the outcome of the research will be will yield inaccurate information as it pertains to the study. The primary weakness of the correlational design is the inability to determine causation.  High correlation cannot be mistaken for causation (Mertens, 2010).  The variables cannot be changed in any way in a correlational design; therefore, researchers have no control over them.  The ability to include several variables within one study is a strength of correlational design but care must be taken not to overload the study with too many variables (Mertens, 2010). 

Correlational Design is Most Useful in an Educational Setting

     When would it be appropriate to use correlation design in an education setting could be answered in numerous ways? Depending on when and how it is used can in many times determine the overall effectiveness.  When researchers want to find and develop the relationships that may exist between variables, they can use correlational designs (Cheng, Yang, Chen, Zou, Su, & Fan, 2016).   When using the correlation design effectively it produces the necessary outcomes that could help researchers conclude the relationships that exist.  If and when the relationships are identified the researcher may be able to successfully manage, the variables and be able to extract the preferred outcome.  Relationships that are created in an educational environment and successful, creates a positive outcome for those that are involved.  An example in correctional design in education could be exploring the relationship between hours that is spent on homework and how well student progress in their course work. Another one could be how well an athlete spends in practice and their success in their sport.

Appropriate or not

     For this study, the correlational design was not a suitable selection for the proposed purpose of the presented study.  The purpose of the study was to describe the relationship concerning the constructs, specifically emotional intelligence, psychological empowerment, resilience, spiritual wellbeing and academic success in undergraduate and graduate nursing students that were involved. During the study what was established was the relationship, but it did not offer a description of it.  The research presented a relationship between the variables used.  Perhaps a more effective approach may have generated better results.  During the analysis, an experimental approach could have been used.  The research could have concluded a more comprehensive look at the relationship that was established with emotional intelligence, psychological empowerment, resilience, spiritual wellbeing, and academic success in graduate and undergraduate nursing students. This study intended to define if there may have been a relationship between emotional intelligence, psychological empowerment, resilience, spiritual well-being, and academic success in both graduate and undergraduate nursing students.  The researcher could not conclude any causality and was unable to control any of the variables that would indeed align well when it comes to the correlational design. 

Types of Problems Correlational Designs

     When a study is trying to achieve the desired outcome, researchers may be challenged with the responsibility to lead them to the best tool and research design that would be most appropriate to use. While trying to achieve this they will have to be engaged and focused on the task. If one wants to show established or discover relationships correlational designs work well (Creswell, 2014).  A strong characteristic of correlational designs it the relationship investigation.  

A Sample Research Question

     Altan, Bektas, Celik, Gerceker, Ok, Ozdemir, & Aricioglu (2018) completed a correlational study. The study that was conducted used a descriptive, correlational, and comparative cross-sectional research design to establish daytime sleepiness and the cause and effect of the study. The researcher that conducted the research use the following question, what is the description of the relationship between emotional intelligence, psychological empowerment, resilience spiritual wellbeing and academic success in undergraduate and graduate nursing students?  For the second selected correlational study, the researcher asked the following question “What is daytime sleepiness is and what are some of the factors that affect it?” 

References

  • Altan, S. S., Bektas, M., Celik, I., Gerceker, G. O., Ok, Y. S., Ozdemir, E. Z., & Aricioglu, A. (2018). Factors Affecting Daytime Sleepiness in Adolescents. International Journal of Caring Sciences11(3), 1840–1848.
  • Beauvais, A., Stewart, J., DeNisco, S., & Beauvais, J. (2014). Factors related to academic success among nursing students: A descriptive correlational research study.
  • Cheng, C., Yang, L., Chen, Y., Zou, H., Su, Y., & Fan, X. (2016). Attributions, future time perspective and career maturity in nursing undergraduates: correlational study design. BMC Medical Education16, 26.
  • Creswell, J. (2014).  Research design: Qualitative, quantitative, and mixed methods approaches.  (4th ed.).  Thousand Oaks, CA: Sage Publications.
  • Creswell, J. W., Creswell, J. D. (2018).  Research Design:  Qualitative, quantitative, and mixed methods. (5th ed.).  Thousand Oaks, CA:  Sage Publications.
  • Fitzgerald SM, Rumrill PD Jr., & Schenker JD. (2004). Correlational designs in rehabilitation research. Journal of Vocational Rehabilitation20(2), 143–150.
  • Mansfield, E. R., & Helms, B. P. (1982). Detecting multicollinearity. The American Statistician36(3a), 158-160.
  • Mertens, D. (2014). Research and evaluation in education and psychology. Integrating diversity with quantitative, qualitative, and mixed methods.  Thousand Oaks, CA: Sage Publications.

Cite This Work

To export a reference to this article please select a referencing stye below:

Reference Copied to Clipboard.
Reference Copied to Clipboard.
Reference Copied to Clipboard.
Reference Copied to Clipboard.
Reference Copied to Clipboard.
Reference Copied to Clipboard.
Reference Copied to Clipboard.

Related Services

View all

DMCA / Removal Request

If you are the original writer of this essay and no longer wish to have your work published on the UKDiss.com website then please:

Related Lectures

Study for free with our range of university lectures!