Short Answer Questions Essay Example

  • Category:
  • Document type:
  • Level:
  • Page:
  • Words:

Short Answer Questions

Question 1

True Experimental and Quasi-Experimental Research Design

Experiments are mostly carried out within a highly controlled environment, for example a laboratory where a random sample of study participants is selected before the beginning of the project. True experiments are normally performed to establish the cause and effect of a project (Campbell & Stanley, 2015). This means that true experiments are conducted to determine if something works or to find a new way of improving the present situation. Generally, true experiments have a validity and reliability to determine the study’s efficacy and genuinity. A true experiment has both experimental group and control group (Campbell & Stanley, 2015). For instance, in a hospital setting patients with liver disease are separated into two subgroups where one group continues using one medication while the other group is administered with the new drug. Researchers then track the progress of the patients for a specific time periods and analyses the outcome to determine the group with better health improvement. In case the new drug is proved to be more effective in treatment, it quickly replaces the previous medication.

On the other hand, quasi-experiments are conducted retrospectively and entail manipulation of the independent variable and there might be comparison of groups. There is no randomised participant selection in quasi-experimental design. Additionally, there is no control over internal validity (Campbell & Stanley, 2015).

Similarities between true experimental designs and a quasi-experimental is that the two designs have an experimental group and a control group. The main difference is that while true-experimental design has a randomised participant selection, there is no randomised participant selection in quasi-experimental design (Campbell & Stanley, 2015).

Question 2

Difference between Internal Validity and External Validity

Internal validity is the validity of the research study itself. This means that internal validity refers to the extent at which the researcher can validate that the study outcome was not affected by other variables other than the variable under study. A study with high internal validity has its outcome directly resulting from the manipulation of the study. Factors that can negatively affect internal validity of a study include; sample selection, bias, attrition, among other factors. For example, a study researching about the variable of self-study and results of examination results should be able to show that no other variables such as IQ, coaching or teaching techniques results to good examination results (Reis & Judd, 2000).

On the other hand, external validity involves generalization of the results of a study to a bigger population. This means that external validity is partly determined by whether it is practical to replicate the study results in and across other samples and settings. A study whose external validity is high can be replicated within several contexts with comparable outcomes. For instance, results of a study performed on a sample population to the population as a whole can be generalised. In the same manner, results of a study performed using sample of some students in a college can be applied to real-world setting such as college (Reis & Judd, 2000).

Generally, the main difference between external validity and internal validity is that in internal validity it is possible o claim that the study outcomes were not influenced by any other variable apart from the variable under study, while in external validity refers to the degree to which the study outcomes can be generalised to the general population (Reis & Judd, 2000).

Question 3 Information provided by Z-Score and the effect of Transformation

A Z-score refers to the numerical measurement of the value’s correlation to the mean within a group of values. A Z-score value of zero means that the score is equivalent to the mean score. The key purpose for a Z-score is identifying and describing location at individual scores within the distribution. Z-score also aims at standardising the entire distribution and can also develop a different distribution corresponding and equivalent (Barbaree & Mewhort, 2004).

Basically, transformation of X values into Z-scores aims at informing the precise position of the original X value in the distribution. The Z-score achieves this by converting every X value into signed number (+ or -) where the sign informs if the location of the score is above or below the mean. The number informs the distance between the score and the mean in form of standard deviation. In addition, the Z-scores also form a standardised distribution that is directly comparable to other distributions that have also been converted into Z-score (Barbaree & Mewhort, 2004).

When the mean, the standard deviation, and the shape of a distribution scores are transformed into z-scores, the following happens:

  • The shape of the distribution scores remains the same as the primary distribution of raw scores

  • The Z-score distribution at all times have a mean of zero

  • The distribution of z-scores will at all times have a standard deviation of one (Barbaree & Mewhort, 2004).

Question 4

Mean Standard Error’ and why it Decreases As Sample Size Increases

The mean standard error estimates the unpredictability between sample means that can be obtained if multiple samples were taken from the same population. For instance, if the mean delivery time is 3.8 days where the standard deviation is 1.43 and this is based on random sample of 312 delivery times. The numbers result to mean standard error of 0.08 days. In case the sample has been obtained from several random sample containing similar size and equivalent population, the standard deviation of the varying samples can approximately be 0.08 days (Gravetter & Wallnau, 2009).

The mean standard error decreases with increase with sample size because with the increase of the sample size, the sampling distributions move toward a normal distribution. Additionally, with the increase of the sample size, the unpredictability of every sampling distribution reduces. The range of the sampling distribution also decreased when compared to the range of the primary population (Gravetter & Wallnau, 2009). The standard deviation of all sampling distributions is equivalent to s/ÖN (N is the sample size obtained from the general population).

Generally, mean standard error establishes how accurately the mean of the sample approximates the population mean. Lower values of the mean standard error show that there is increased accuracy of the population mean estimates. Normally, a higher standard deviation results to higher mean standard error and reduced accuracy estimate. A higher sample size results to lower mean standard error and a more accurate estimate (Gravetter & Wallnau, 2009).


Barbaree H & Mewhort D. (2004). The effects of the z-score transformation on measures of relative erectile response strength: a re-appraisal. Behav Res Ther. 32(5), pp:547-58.

Campbell D & Stanley J. (2015). Experimental and Quasi-Experimental Designs for Research. New York: Ravenio Books.

Gravetter F & Wallnau L. (2009). Statistics for the Behavioral Sciences. London: Cengage Learning.

Reis H & Judd C. (2000). Handbook of Research Methods in Social and Personality Psychology. London: Cambridge University Press.