China
Africa
Explore chapters and articles related to this topic
Inference from probability and nonprobability samples
Published in Uwe Engel, Anabel Quan-Haase, Sunny Xun Liu, Lars Lyberg, Handbook of Computational Social Science, Volume 2, 2021
Rebecca Andridge, Richard Valliant
In network sampling, members of some target population (usually a relatively rare one like intravenous drug users or men who have sex with men) are asked to identify other members of the population with whom they are somehow connected. Members of the population that are identified in this way are then asked to join the sample. This method of recruitment may proceed for several rounds. Snowball sampling (also called chain sampling, chain-referral sampling, or referral sampling) is an example of network sampling in which existing study subjects recruit additional subjects from among their acquaintances. These samples typically do not represent any well-defined target population, although they can be a way to accumulate a sizeable collection of units from a rare population.
Social value assessment
Published in Ani Raiden, Martin Loosemore, Andrew King, Chris Gorse, Social Value in Construction, 2018
Ani Raiden, Martin Loosemore, Andrew King, Chris Gorse
Whilst probability-based (random) sampling can save time and expense, most researchers do not have the resources to randomly sample an entire population and therefore find it necessary to employ non-probability sampling. In non-probability sampling, respondents are selected on the basis of a purposive personal judgement of the researcher. This can be done in a number of ways. For example, convenience sampling involves picking your sample according to what is available. Snowball sampling involves a researcher asking one target person in the population to nominate other potential participants based on their knowledge so that the sample gradually snowballs to become larger. Judgement sampling involves a researcher deciding which population members to include based on his or her judgement, which is justified in some way. Quota sampling includes a designated number of people with certain specified characteristics. The downside of non-probability sampling is that the choice can be biased by the researcher’s preferences and perceptions, and an unknown proportion of the entire population is not sampled, which means that the sample may or may not represent the entire population accurately and, as such, the results of the research cannot be generalised to the entire population.
An Ethnographic Approach to Design
Published in Julie A. Jacko, The Human–Computer Interaction Handbook, 2012
Jeanette Blomberg, Mark Burrell
Snowball sampling relies on participants referring others whom they think would be good candidates for the research, or on researchers identifying individuals or groups to be included in the study as the research proceeds. Because this method utilizes existing social networks it is especially valuable when desired participants are initially inaccessible or reluctant to participate (e.g., CEOs, drug users, club members) or when the relevant population cannot be known in advance.¶
Mobile 3D body scanning applications: a review of contact-free AI body measuring solutions for apparel
Published in The Journal of The Textile Institute, 2023
Sadia Idrees, Simeon Gill, Gianpaolo Vignali
In the first phase, the applications were identified writing keywords ‘3D body scanning mobile applications, 3D body scanner, 3D body scanning application’. The keyword searches and identified applications names further lead to identifying various 3D body scanning apps. The data was obtained from peripheral databases sources such as Google, Google scholar, App store, Google play and academic publication for initial search of 3D body scanning apps. The sampling method adopted for data collection was snowball sampling. Snowball sampling is a non-probability sampling method. In this sampling technique the existing subjects provide referrals for recruitment of identical samples needed for a research study (Goodman, 1961). Therefore, this sampling technique has been employed to find out the existing 3D Body scanning mobile applications through online secondary data sources. Secondary data is the data compiled from prior studies, interfaces and books (Ghauri & Gronhaug, 2010). Secondary data is essential as it supports identification of the research problem and describing the research questions, assisting to formulate project, elucidate the data, provide awareness and validate the conclusions (Malhotra et al., 2012).
Comparing human and cognitive assistant facilitated brainstorming sessions
Published in Journal of Engineering Design, 2022
Torsten Maier, Nicolas F. Soria Zurita, Elizabeth Starkey, Daniel Spillane, Christopher McComb, Jessica Menold
28 participants (9 teams) participated in this study. 5 teams completed the study with the assistance of the CA facilitator and 4 teams with the human facilitator. Sessions were ordered based on participant’s schedule and availability. Additionally, participants were only allowed to participate in one session. The average age was 24.86±3.17. 18 students were male and 10 were female. Additionally, all participants were graduate students at The Pennsylvania State University; 61% were pursuing degrees in Industrial Engineering, 25% in Mechanical Engineering, 11% were in Engineering Design, and 4% were in Architecture. Participants were recruited through campus resources (e.g. societal meetings, email list serves, etc.). Individuals that completed the study were then invited to recruit future participants, a recruitment technique known as purposeful snowball sampling (Naderifar, Goli, and Ghaljaie 2010). Purposeful snowball sampling was used because it allows researchers to expand their recruitment pool beyond those participants that they have direct access to. Prior to the study, all participants were informed that their participation was voluntary, and all information would remain confidential, following Institutional Review Board policy. Although relatively few teams were recruited for this study, a wealth of data (thousands of samples) were collected for each of those teams, allowing the researchers to make use of rigorous statistical models. For instance, the audio/video recordings of the brainstorming sessions were coded using a timestep of 0.2 s resulting in 4516 codes on average per team.
Understanding the emergence of redistributed manufacturing: an ambidexterity perspective
Published in Production Planning & Control, 2019
Samuel Roscoe, Constantin Blome
Our sampling logic was to conduct preliminary interviews with key experts at a UK pharmaceutical policy body. Drawing on their in-depth knowledge about the UK pharmaceutical industry, these experts helped identify case companies that could assist in answering the research question. Based on their advice and to allow comparisons according to firm size, we studied five MNEs and five SMEs. The policy experts introduced us to key contacts at the case companies and we then used a snowball sampling logic to identify potential interviewees. Snowball sampling is a non-probability sampling technique in which existing subjects recruit future subjects from among acquaintances in their social network until the desired sample size is reached (Morgan 2008). A potential drawback of convenience sampling is that the sample may not be representative of the population. However, because of confidentiality concerns, the researchers found that the only way to gain access to prospective interviewees was through fellow employees at each case company.