China
Africa
Explore chapters and articles related to this topic
Sampling Theory
Published in Marcello Pagano, Kimberlee Gauvreau, Heather Mattie, Principles of Biostatistics, 2022
Marcello Pagano, Kimberlee Gauvreau, Heather Mattie
Unlike simple random sampling, systematic sampling requires the selection of only a single random number – a random number between 1 and k. Also, it distributes the sample evenly over the entire sampling frame. Bias may arise if there is some type of periodic or cyclic sequence to the list. For example, if k is an even integer, and if the numbers on the list are sequential, then those in the sample will all be even or odd, depending on the first number chosen. This may mean we choose houses on one side of the street only, but such patterns are rare. If we can assume that the list is randomly ordered, then each individual has an equal chance of being chosen. However, this does not mean we have a simple random sample. Not every sample of size n has an equal chance of being chosen. To see this, consider the first step where we choose a number between 1 and k. There are only k possible samples we can choose – far less than the number of possible simple random samples. This is an example of a cluster sampling design with a single cluster. Cluster designs are addressed in Section 21.1.4.
Social Class for Collective Health Research: A Conceptual and Empirical Challenge
Published in Emily E. Vasquez, Amaya Perez-Brume, Richard G. Parker, Social Inequities and Contemporary Struggles for Collective Health in Latin America, 2020
Adriana Gisela Martinez-Parra, César Ernesto Abadía-Barrero, Chiharu Murata, Ignacio Méndez Ramírez, Ignacio Méndez Gómez-Humaran
ENSAB IV is a cross sectional study with a stratified and multistage sampling design. The sample size of 20.533 people represented the six geographical regions most commonly used for official surveys. The sample included people belonging to the following age groups 1, 3, 5, 12, 15 and 18; adults between the ages of 20 and 79; and pregnant women between the ages of 20 and 49.
Measurement of Exposure and Dose
Published in Samuel C. Morris, Cancer Risk Assessment, 2020
Measurement errors of physical and chemical properties of water contaminants are generally well within 25% of observed mean values; natural variations over time or space are much larger, typically 100% to 400% (Mar et al., 1986). This illustrates the importance of sampling design, an area sometimes given less attention than laboratory analysis.
Participatory research on the prevalence of Multi-Substance vaping in college students
Published in Journal of American College Health, 2023
A non-probability sampling design was employed after discussion in the class about the strengths and limitations of representative vs. non-representative samples. Students led and tracked administration of the survey tool in a shared Google document. Students disseminated the survey electronically to all Deans of colleges at the University, posted the survey on a campus-wide social media platform, and on the Facebook pages of student organizations including cultural groups, sororities and fraternities, and one sports team. One student reviewed the course catalog to identify nine large courses of 100–300 students in each discipline and emailed the instructors of each course to request that they distribute the survey to their students. Many students in the course also emailed peers in their own courses via the course learning management software requesting that they complete the survey. The survey was opened on March 25, 2019 and closed on April 15, 2019. The survey was anonymous and participants could choose to opt out after reading a notice of consent.
Past-Year Ketamine Use: Evidence from a United States Population, 2015-2019
Published in Journal of Psychoactive Drugs, 2023
Pooled data from the 2015–2019 National Survey on Drug Use and Health (NSDUH) were analyzed. The NSDUH is a cross-sectional, annual survey conducted in the US to measure health behavior prevalence, substance use, and behavioral health utilization among individuals 12 years or older. A complex sampling design was used to ensure adequate response rates and equal probability selection of individuals. Other details of the NSDUH are explained elsewhere (Center for Behavioral Health Statistics and Quality 2019). The response rates for the 2015–2019 NSDUH were between 67–73%. I used data from 2015 on because that was the first year NSDUH asked about sexual identity, and previous research has identified sexual identity as a risk factor for ketamine use (Roxburgh et al., 2016). For the present analyses, I excluded individuals who reported their sexual identity as “Don’t Know” (n = 1,364) and “Refused” (n = 2,630), making the final analytic sample 210,392 individuals 18 years or older. The University of North Texas Health Science Center Institutional Review Board approved this as non-human subjects research, and therefore excluded from internal review.
Prevalence of cardiovascular disease risk factors in Chinese patients with type 2 diabetes mellitus, 2013–2018
Published in Current Medical Research and Opinion, 2022
ChenChen Wang, ZuoLing Xie, Xi Huang, Zheng Wang, HaiYan ShangGuan, ShaoHua Wang
The present cross-sectional study included consecutively all patients hospitalized in the Zhongda Hospital affiliated to Southeast University, whose medical records had diagnoses with diabetes, from July 2013 to the end of 2018. This study used a cluster sampling design. A total of 66,536 cases of DM met the criteria for study inclusion. Excluded from the study were 5411 cases due to type 1 diabetes diagnosis, gestational diabetes, specific type diabetes or diabetes not clearly classified; 30,013 cases due to records of multiple hospitalizations; and 419 cases due to severely missing data. A flowchart of the study procedure based on our previous survey16 is shown in Figure 1. The procedure of this study was approved by the Research Ethics Committee of Zhongda Hospital affiliated to Southeast University (Approved no. of ethic committee: 2020ZDSYLL028-P01).