The Level of Measurement for Variables - Coursework Example

The Level of Measurement for Variables Question: How does knowing the level of measurement for variables help a researcher examine outcomes in a study? What is the relationship of precision and levels of measurement? Answer: Level of Measurement links a value or a category of a variable to some attributes of the observed case. Knowing the level of measurement is very important because it allows the researcher to have a clear understanding of variation in the variables. At the same time, the researcher can also have a better command over the interpretation of these variables. When the researchers have appropriately identified the levels of measurements, they are in a better position to accurately select and use the statistical tools to examine the outcomes of the study. This is an essential step because each level of measurement of a particular variable shows different kind of information and therefore, puts limitation on the use of several statistical and inferential tools. When the researcher, on the other hand, wants to analyze the relationship between different variables, it will also be better placed to identify the necessary statistical tools. For example: A researcher at a university was concerned with analyzing the association of the student gender and the major field of study. Since both the variables are nominal, the researcher can most suitably use the Chi-Square test (Ahmed, 2009). Therefore, knowing the levels of measurements for variables always help a researcher in examining the outcomes of the study more appropriately with the most suitable use of the statistical, inferential and analytical tools. Precision of the data collected for a particular variable is a big concern for the levels of measurement. Basically, it defines how finely calibrated and precise the scale should be. At the same time, the precision determines the usage of the most appropriate statistical tools and their interpretation (Ahmed, 2009). Question: Discuss the difference between random sampling and stratified random sampling. When is it most appropriate to use stratified random sampling techniques? Answer: In a random sampling method, each and every element of the population has an equal chance of being selected. This ensures that the chosen element is the true unbiased representative of the population. On the other hand, a stratified random sampling method first divides the population into small groups known as strata. These divisions are based on the members of the populations having similar characteristics. A proportionate random sample is then taken from each strata and pooled together to create a random sample. The main difference is that in stratified random sampling sub-groups are created in the overall population and proportionate sample are taken whereas in random sampling no groups are created. Also, the stratified sampling is more accurate representative of the true population and therefore less sampling error. The stratified random sampling technique is most appropriate to use when the population is diverse with a variety of attributes and varies immensely within the region. This allows outcomes to be made with great accuracy and comparisons to be done between the sub-groups. It is more suitable to use when the researcher wants to attain precise statistical significance in a smaller sample and wants to reduce the sampling error. An example of this sampling is the US political survey in which the groups are made based on various attributes such as race, religion and location. Question: When might a quasi-experimental design be chosen over an experimental design? What considerations must a researcher make when deciding which approach to take? What are the threats to validity and reliability is this design trying to control for? Answer: A Quasi-experiment design is preferred over an experiment design when the researcher is unable to control all the factors that influence the outcome of the experiment. The people cannot be randomly assigned to the treatment because it tends to sometimes unethical or impractical or expensive or the researchers have a particular interest in the intact group. For Example: A public health program. They can still find out the causal relationship but have to put in an extra effort. Basically, a quasi-experiment is performed when there will be: (Shuttleworth, 2008) No proper stated hypothesis for the causal relationship No control group No randomization of the control and treatment groups No treatment of the independent variable. The scientists have regarded this method as unscientific. These experiments basically lack the factor of randomization and therefore, good statistical an1alysis cannot be performed. These statistical tests are meaningless because the researcher did not control the other factors affecting the experiment. Hence, when the researchers consider the quasi-experiment design, the reliability of the experiment is lost. At the same time, there are some threats to the internal validity present in the quasi-experiment: (Shadish, 2002) Selection bias: At the end of the experiment, the people tend to be different and those effects are also attributed to the treatment. Example: Self Selection History effects: Many other events also occur during the course of the experiment. Example: Strikes Testing : People with extreme scores on one assessment tend to move towards the mean score in the latter assessments because of their previous experiences Differential Mortality: Difference in the drop outs between the two treatment groups Question: What are some examples of pre-experimental research which you observe in everyday life? Answer: Any set of observations concluded on a particular behavior is defined as the pre-experimental research. There are many examples of pre-experimental research that one can observe in everyday life. One of the most appropriate examples is the rise of the stock prices before a holiday. Anyhow, in pre-experimental design, there are three main structures: (Heffner, 2004) First is that the subjects are exposed to treatment and the outcome is measured. Example: In any university across the world, after a semester of heavy work, the grades are observed and measured. Second is that the subjects are pretested for their behavior. They are then exposed to a treatment and post tested for that behavior. Any deviation is the result of the treatment. Example is the improvement in the college grades. In this case, we can compare the grades prior to experience to the ones post a semester worth of experience. Third is the two different groups exposed to the same treatment. Example is that the boys sitting next to girls have lower grades. There are basically two groups of boys. One is made to sit with the girls and the other with the boys. The outcome of the grades was measured after a semester and proved that the boys sitting next to girls have poor grades. Question: Describe how a research study should be evaluated. Give examples of your comments of the evaluation. Answer: A research study should be intensively evaluated because it answers an intriguing question and will become the basis for future researches. The study should be lucid and focused signifying the researcher’s knowledge and perceptions. At the same time, the order, structure and presentation of the paper should be coherent and accurate so that the reader finds it interesting to study. The study should also be evaluated on the basis that: The researcher has used authentic sources to verify the points. Example: Any book, journal or previous research has been used then proper referencing should be done. The researcher is not heavily dependent on one source but rather has a variety of different sources. Example: There are more than one sources stated in the reference list which means the work is validated. The research is a pure blend of the researcher’s points as well as the documented ideas. Example: The current research is supported by the previous researches and is not baseless. Appropriate documentation is included. Example: The referencing is appropriately done following the APA referencing styles for books, journals, online pages and periodicals Quoted material is properly certified and the documented The documentation is precise. Example: There should be no errors in the documentation because it will be the basis of your current research. Question: List some of the rules for APA style governing reference lists. Explain why the APA style is used in research proposal writing. Write a checklist to use when proofing a project for adherence to APA style. Answer: American Psychological Association (APA) is the most commonly used citation style in the social sciences. The referencing section starts on the new page with the word “Reference” centered on the first line. Some of the basic rules for APA style governing reference lists are: (OWL, 2009) Hanging indentation is to be present which means that all lines after the first line of each reference entry must be indented one-and-half inch from the left margin. Names of the authors are inverted with a maximum of six that can be only mentioned. If there are more than six authors et al is included at the end Entries are arranged alphabetically by the last name of the first author of each entry If there are multiple articles by the same author, these references are then arranged according to the publication date, starting with the earliest. The research proposal writing concentrate on a particular project: academic or scientific research. The APA style simplifies the work and allows the reader to easily understand the text because it follows a similar pattern. The proposal contains intensive literature reviews that would provide convincing support of need for the proposed research. Therefore, to authenticate these literature reviews and to help the panel of experts to locate these resources: the APA referencing style is used (OWL, 2009). Checklist to be used when proofing a project for adherence to APA style: Typed Double Spaced on a standard sized page Time New Roman 12 point 1” margins on all sides Page Header with the “Title of your Paper” Headings are bolded and centered Sub-headings are in italics and on the left side The first line of the paragraph is indented one-and-half inches References Ahmad, T. (2009, January 22). Understanding Levels of Measurement. Retrieved February 3, 2010, from http://ezinearticles.com/?id=1910066 Heffner, C.L., (2004, March 11). Research methods. Retrieved on Feb 3, 2010 from http://allpsych.com/researchmethods/preexperimentaldesign.html Purdue University Online Writing Lab (OWL) (2009). APA General Format. Retrieved Feb 3, 2010, from http://owl.english.purdue.edu/owl/resource/560/01/ Salkind, N.J. (2008). Exploring Research. Prentice Hall. Shadish, W.R., Cook, T.D., & Campbell, D.T. (2002). Experimental and quasi-experimental designs for general causal inference. Boston: Houghton Mifflin Shuttleworth, M. (2008). Quasi-Experimental Design. Retrieved on Feb 4, 2010 from http://www.experiment-resources.com/quasi-experimental-design.html Read More