As with most of our lessons in this class, the difference between descriptive and inferential statistics are fundamentally different. Also, the two types of statistics are inherently more beneficial to either qualitative or quantitative research methods respectively.
Descriptive statistics are meant to provide an overview or summary, often visually, of samples and measurements of a particular study. (APUS 2016) Descriptive statistics, unlike inferential statistics, are not meant to be used to formulate conclusions from. They are meant to solely give the readers an overview of the findings. While ideally suited for quantitative studies, descriptive statistics can still be used to visually represent data collected in a qualitative study. Descriptive statistics can also be used to measure central tendency. This is done by using three measurements, the mode, median and mean. The mode is the central position when all numerical datasets have been added together and then divided by the number of datasets. The median is the center of the dataset when all results are laid out from least to greatest. Lastly, the mode is the number that appears most often in the collected data. (APUS 2016)
In comparison, inferential statistics is generally what qualitative studies use to draw conclusions from. Inferential statistics allow the research to formulate an explanation and draw conclusions based on the data that was collected in their research. This type of statistic lines up perfectly with qualitative studies because the nature of qualitative studies and the data that is collected is often not conducive to formulate an opinion based solely on raw data. The forming of a narrative from raw data is often the point of a qualitative study, which is why this statistical analysis method is the go to for qualitative studies. For the research proposal that I elected to propose in this class, inferential statistics would be the statistical method that I would employ. A survey of a population with five questions asking for opinionated data, which will be used to form a narrative, describes inferential statistics perfectly. Hence, this is the method I would use.
P-Values are what a researcher uses to compare to their set significance level. The significance level of a test must be set before the testing is carried out by the researcher. The significance level is represented by the letter “A” or alpha. Significance level can be described as in the simplest of terms, is the threshold probability of incorrectly rejecting the null hypothesis when it is in fact true (a type I error). (APUS 2016) This relationship is used to determine if a null hypothesis is true or not. If the p-value is less than or equal to the alpha, the null hypothesis is rejected. (APUS 2016) Conversely, if the p-value is greater than the alpha, the will have failed to show the need to reject the null hypothesis. This is important in quantitative research because data is easily categorized, which makes it simple to determine if the p-values render the null hypothesis significant or not. (APUS 2016)
I enjoyed talking to everyone throughout the last 8 weeks and wish everyone good luck in their future endeavors. Take care!
American Public University. (2016). Understanding Research Results: What is Next? Retrieved January 21, 2019, from https://edge.apus.edu/access/content/group/security-and-global-studies-common/Universal/SSGS/300/elf/lesson-8/elf_index.html
This is the last week and can’t wait to get it over I only have three more classes to graduate. Descriptive and inferential statistics forms the two key branches of statistics science. Descriptive statistics is the term used in the analysis of data that aids describe, show or condense data in a significant way such that, for instance, patterns might arise from the data., however, descriptive statistics do not allow a researcher to make conclusions beyond the data he or she has analyzed or arrive at conclusions concerning any hypotheses that might have been made (Crossman 124). Descriptive statistics offer a brief summary of data. Descriptive statistics therefore enables the researcher to present the data in a more meaningful approach, which permits simpler elucidation of the dataInferential statistics apply a random sample of data taken from a population to describe and make inferences about the population. Inferential statistics are used when it is not convenient or impossible to examine each member of a whole population. For instance, it is impossible to measure the diameter of each nail that is manufactured in a mill, but one can measure the diameters of a representative random sample of nails and use that information to make generalities about the diameters of all the nails formed (Crossman 145).Data analysis in qualitative research is always analyzed thematically and not statistically, that is the reason why descriptive and inferential statistics is not an issue. The analyzed data is usually reported in form of narratives with first hand quotation from primary sources. Before conducting the analysis, the researcher ought to decide on a level of significance or a P value that to be used as a limit for accepting that a statistically significant result indicates true differences between groups. The P value reveals the probability that the statistical result can ensue by chance, and it establishes the risk of the researcher making a type I error (Loannidis 170). This implies that a null hypothesis is forbidden when in reality the hypothesis is true, leading to an incorrect interpretation that an intervention was successful. Equally, a type II error is the acceptance of a null hypothesis when it is actually false; in this situation, an intervention that is successful is not recognized as such. Statistical procedures are suitable for analyzing data from complex correlational and experimental studies that have several independent and/or several dependent variables. Usually, large sample sizes are required to accommodate analysis of increasing numbers of variables. The procedures commonly used comprise of multiple regressions, path analysis, and analysis of covariance (ANCOVA), factor analysis, discriminate analysis, canonical correlation, and multivariate analysis of variance (MANOVA). As research becomes more sophisticated, the use of multivariate statistics in studies increases. This is a dilemma for first time researchers and research consumers because studies using complex analytic procedures may be more difficult to assess (Loannidis 172). Results, Discussion, Conclusions, and Dissemination Study results or findings should be clear, concise, and congruent with the research question(s) asked. Good luck class and best wishes to you.
Crossman, A. “Descriptive vs. inferential statistics.” About. com (2013).
Ioannidis, John PA, et al. “Increasing value and reducing waste in research design, conduct, and analysis.” The Lancet 383.9912 (2014): 166-175.