China
Africa
Explore chapters and articles related to this topic
Standardizing an Assessment
Published in Lucy Jane Miller, Developing Norm-Referenced Standardized Tests, 2020
James Gyurke, Aurelio Prifitera
Stratified sampling is a controlled selection for certain known characteristics of the population.3 This method is the most commonly used when standardizing tests where the goal is to collect a national normative sample reflecting certain characteristics of the U.S. population which are thought to influence performance on the assessment being standardized. For example, a stratified sample will consist of a predetermined proportion of white males, between the ages of 10 and 12, who are middle class and reside in the eastern United States. After the stratification cells have been determined, the filling of each cell is done using a random process. An example of a stratification chart is presented in Table 4.
Sampling and Sample Size
Published in Abhaya Indrayan, Research Methods for Medical Graduates, 2019
In descriptive studies, sample mean and sample proportion do provide a reasonable idea of the mean and proportion in the corresponding target population. As usual, this “reasonability” is assessed in terms of validity and reliability. When the sample subjects represent the full spectrum of a population, which is likely if one of the random methods described earlier is followed and if the sample size is reasonably large, sample mean and sample proportion are indeed valid point estimates. That is, they are unbiased in the sense of being able to reach the corresponding population value when the sample size is increased, and fairly stable across samples. This statement has two underlying assumptions. First, the mean or proportion is calculated after due consideration of varying group size, as can happen in the case of stratified sampling. Second, the data obtained from the sample subjects are correct, that is, no wrong data are reported or recorded.
Sampling Theory and Methods
Published in M. Venkataswamy Reddy, Statistical Methods in Psychiatry Research and SPSS, 2019
Stratified sampling may be preferred to increase the precision of the estimate by reducing the heterogeneity of the population. In stratified sampling, the population of N units is divided into L sub-populations of N1, N2, … NL units that are internally homogeneous. These sub-populations are not overlapping, and together they comprise the whole population. The classes into which the population is divided are called strata. Then, each stratum will be sub-sampled, and a definite number of units are taken from each stratum as in the case of simple random sampling. The partial samples of the different strata may be used to estimate the means of several strata from which the overall estimate of the population mean may be obtained.
Contributions of safety critical success factors and safety program elements to overall project success
Published in International Journal of Occupational Safety and Ergonomics, 2023
Mohanad Kamil Buniya, Idris Othman, Riza Yosia Sunindijo, Ali Amer Karakhan, Ahmed Farouk Kineber, Serdar Durdyev
The sampling method chosen for this study was stratified sampling derived mainly by the fact that the focus on worker safety is a relatively recent development in the Iraqi construction industry [2]. This sampling approach is also considered as an effective method in conducting research on safety in the building sector [96–98]. This sampling method is related to how a specific population is selected as a sample [99]. Even though the safety program in Iraq is relatively new, stratified samples were used to collect data from a specific sub-population [100]. This approach assisted the researchers in acquiring the most accurate and reliable data. The benefits of stratified sampling as highlighted by Sharma [101] are as follows: (a) decreasing bias in sample case selection, implying that the sample can represent the surveyed populations; (b) allowing the sample to be generalized to the population. In this case, the population difference is considered by stratification, along with all three sectors (client, contractor and consultant) and five subsectors in Iraq [102].
From COVID-19 Vaccination Intention to Actual Vaccine Uptake: A Longitudinal Study Among Chinese Adults After Six Months of a National Vaccination Campaign
Published in Expert Review of Vaccines, 2022
Jiahao Wang, He Zhu, Xiaozhen Lai, Haijun Zhang, Yingzhe Huang, Huangyufei Feng, Yun Lyu, Rize Jing, Jia Guo, Hai Fang
Our study has several limitations. First, this study used an online survey design and the sample may not be well representative. The stratified sampling was used to reduce the limit of the sample representativeness. Second, due to the self-reported nature of the study, reporting bias may exist. Though only some of the participants, those with prior intentions especially for acceptors, were asked in recall information in the follow-up survey, and they were distributed in both groups who with and without prior vaccination, the possible bias is worth noticing when interpreting results. Third, due to the comprehensiveness of interventions adopted in China’s vaccination campaign, it may be unfeasible to differentiate the effects of each intervention, so we examined them as a whole. Fourth, the cross-sectional nature of the regressions may not inform a causal interpretation of the predictors, as the dependent variable, whether each individual accepted the vaccination, does not have a longitudinal characteristic. To the best of our knowledge, this is the first study to assess the intention-action gap of COVID-19 vaccination, with the advantage of a cohort sample, and to examine the mid-term impact and shortcomings of the vaccination campaign in China for the first year.
The Prevalence of Sexual Abuse in Adolescence in Suriname
Published in Journal of Child Sexual Abuse, 2019
Inger W. van der Kooij, Shandra Bipat, Josta Nieuwendam, Ramón J.L. Lindauer, Tobi L.G. Graafsma
Sampling was done through a multistage stratified sampling technique. A list of schools for boys and girls, arranged by type of school, grades taught, and numbers of boys and girls, was accessed through the Inspection of Education (Nickerie) and the Ministry of Education and Development (Paramaribo, Marowijne, Sipaliwini and Brokopondo). Based on student populations in each of the five selected regions of the country, a weighted student sample size from each of these regions was identified. Geographical boundaries within each of the major districts were identified to ensure that demographic differences in the districts were addressed. In addition to district, school type (level and religious background) and year, gender and the size of the participating school were considered. Data from the study in Nickerie was used to enhance the study population and its cultural diversity and thereby improve the representativeness of the sample. As both studies used the same questionnaire, the data from the studies could be combined.