China
Africa
Explore chapters and articles related to this topic
Introductory Statistics Refresher
Published in Jiro Nagatomi, Eno Essien Ebong, Mechanobiology Handbook, 2018
Julia L. Sharp, Patrick D. Gerard
An observational study can be thought of as a relatively passive study where measurements are observed but there is no attempt to influence the outcome. In these studies, no treatment is being imposed and only existing conditions or outcomes are under study. The sample in these types of studies should be carefully selected so that results can be generalizable to the population of interest. Several popular sampling designs include the simple random sample, the stratified random sample, and the cluster sample. To obtain a simple random sample, a group of individuals, units, or subjects is selected so that every possible sample of the same size has the same chance of being selected. A stratified random sample is selected by dividing the group of individuals, units, or subjects into separate groups or strata. A simple random sample is selected from each strata. In a cluster sample strategy, a simple random sample of groups or clusters is selected and then all individuals within the selected group are selected. Statistical methods have been developed to consider objectives from each of these types of sampling schemes.
Inference
Published in Julian J. Faraway, Linear Models with Python, 2021
For observational studies, we envisage a finite population from which we draw the sample that is our data. We want to say something about the unknown population value of β, using estimated values β^ that are obtained from the sample data. We prefer that the data be a simple random sample of the population. We also assume that the size of the sample is a small fraction of the population size. We can also accommodate more complex random sampling designs, but this would require more complex inferential methods.
Sampling—Measurement Variables
Published in Frank R. Spellman, Fundamentals of Wastewater-Based Epidemiology, 2021
In stratified sampling, a population is divided into subpopulations (strata) of known size, and a simple random sample of at least two units is selected in each subpopulation. This approach has several advantages. For one thing, if there is more variation between subpopulations than within them, the estimate of the population mean will be more precise than that given by a simple random sample of the same size. Also, it may be desirable to have separate estimates for each subpopulation (e.g., administrative subunits). And it may be administratively more efficient to sample by subpopulations.
When does control curb opportunistic behaviour: evidence from the construction industry
Published in Production Planning & Control, 2023
Yinqiu Tang, Yongqiang Chen, Hongjiang Yao, Yuting Chen
Since a truly simple random sample of recent construction projects is almost infeasible in the construction industry (Franz et al. 2017) and often results in poor quality due to limited direct access to target respondents (Zhang and Qian 2017), the data in this study were collected using the convenience sampling method. A total of 366 questionnaires were obtained during the four months of data collection from employee training programs in China. Those vocational educations were administrated by construction companies or industry associations to improve the project management skills of construction project professionals. Note that we did not employ a random sampling strategy because of the unit of analysis, i.e. construction projects, it is difficult to identify the clear population of sampling. At the same time, compared to stranger respondents, trainees have a greater sense of responsibility to give detailed and accurate answers to the survey questions, which is conducive to ensuring the quality of the survey data. To better ensure the validity of the sample, the questionnaires completed in less than eight minutes (the minimum time required for completion based on the pilot survey) and the questionnaires marked with the same score across most questions were eliminated.
On Combining Independent Tests in Case of Log-Normal Distribution
Published in American Journal of Mathematical and Management Sciences, 2022
We shall assume that the i-th problem in case of the log-normal distribution is based on which are independent rv.’s. By sufficiency, we may assume ni = 1 and for Then we consider the sequence of independent test statistics, that is, we will take a simple random sample of size n and let and compare the six non-parametric methods via exact Bahadur slope (EBS). Although Xi is not sufficient for θi under for the other distributions, but we will assume ni = 1 and for Our justification for this is that we will get new tests in this case beside other known tests.
New mean charts for bivariate asymmetric distributions using different ranked set sampling designs
Published in Quality Technology & Quantitative Management, 2018
Ranked Set Sampling (RSS) design can be described as follows:Select a simple random sample of size units from the target finite population and divide them into n samples each of size n.Rank the units within each sample in increasing magnitude using personal judgement, eye inspection or based on a concomitant variable.Select the ith ranked unit from the ith sampleRepeat steps 1 through 3, m times if needed to obtain a RSS of size .Let