China
Africa
Explore chapters and articles related to this topic
Assessing the relationship between exercise and employee mental health: methodological concerns
Published in John Kerr, Amanda Griffiths, Tom Cox, Workplace Health, Employee Fitness and Exercise, 2020
Steve M. Jex, Deanne A. Heinisch
A quasi-experiment is a design which is similar to a true experiment, but differs in that one or more of the characteristics of a true experiment is missing. For example, like the example of a true experiment given above, a quasi-experiment may contain a group which participates in a fitness programme, and one which does not. However, in a quasi-experiment, the basis for inclusion into each of these groups may not be random. In many cases, it will simply be on the basis of self-selection. That is, some employees will choose to participate in the fitness programme while others will not. Another crucial difference is that in a quasi-experiment, the independent variable (which, in this case, would be a fitness programme) is usually not manipulated by the researcher or created for research purposes. Organisations usually decide to implement an employee fitness programme for reasons other than research (e.g. to lower health care costs, to provide an additional benefit). As one might expect, quasi-experimental designs are used much more often than true experiments when organisations evaluate fitness programmes (cf. Falkenberg, 1987).
Meso Applications of Wastewater Analysis
Published in Jeremy Prichard, Wayne Hall, Paul Kirkbride, Jake O’Brien, Wastewater Analysis for Substance Abuse Monitoring and Policy Development, 2020
Jeremy Prichard, Wayne Hall, Paul Kirkbride, Jake O’Brien
Opportunities for quasi-experimental designs arise when conditions change, for example because of the implementation of a new strategy, or because of unplanned events or factors. Drug use in the area or group subject to the change can be compared with that in similar groups or areas where no such change has occurred. Such designs can provide useful information for policy evaluation and causal inference, although quasi-experiments cannot control extraneous variables (e.g. selection bias) as effectively as RCTs. Hall (2018) has argued that in the context of substance use, analyses of the effects of changes in conditions tend to be retrospective and sometimes limited because they draw on data that were collected for different purposes, such as ‘event data’ gathered by health and law enforcement agencies (see 1.3.1).
Using Quasi-Experimentation to Gather Design Information for Intelligent Tutoring Systems
Published in Charles P. Bloom, R. Bowen Loftin, Facilitating the Development and Use of Interactive Learning Environments, 2020
A. Scott Wolff, Charles P. Bloom, Anoosh K. Shahidi, Robert E. Rehder
An investigator may choose to implement a quasi-experimental design over other types of designs for a variety of reasons such as: Constraints of the investigator’s operating environment do not permit complete, randomized designs (as discussed earlier).The investigator may wish to explicitly avoid the intrusive manipulations and control so as to reduce the effects of observation as much as possible (as in ethnographic studies).The environment may not permit a well-controlled design (e.g., in a naturalistic setting such as a classroom).
Evaluating digital mathematical games in improving the basic mathematical skills of university students
Published in International Journal of Mathematical Education in Science and Technology, 2022
Marivel Go, Rodolfo Golbin, Severina Velos, Johnry Dayupay, Wym Dionaldo, Feliciana Cababat, Miriam Bongo, Christos Troussas, Lanndon Ocampo
Rather than using a control group, a quasi-experimental method with a single group design is used since students are admitted to different programmes with different courses and instructors. A survey design is also implemented via a pre-and post-test instrument administered before and after the intervention. Also, the attitude of students is measured using a survey instrument. Note that a quasi-experimental design includes a non-equivalent control group, interrupted time series, and stepped wedge designs (Betrán et al., 2018; Brown & Lilford, 2006; Hussey & Hughes, 2007). Such stepped wedges specifically posit that all participants receive the intervention in a staggered fashion. Furthermore, quasi-experimental designs, which are not unique to implementation science, can satisfactorily serve their purpose, especially in cases when experimental designs would become inappropriate (Miller et al., 2020) due to, for instance, in the case of this work, the difference of programmes, topics, and assigned instructors for each course. Using controlled groups in such a scenario threatens the correctness of the results and their corresponding interpretations.
The effect of computer-supported collaborative learning using GeoGebra software on 11th grade students’ mathematics achievement in exponential and logarithmic functions
Published in International Journal of Mathematical Education in Science and Technology, 2022
In this study, a quasi-experimental research design with a pre-test and a or practical to control all the key factorspost-test was conducted. In the true experiments research design, all the important factors that might affect the phenomena of interest are completely controlled. In a true experiment, participants are randomly assigned either to the treatment or the control group, whereas they are not assigned randomly in a quasi-experiment research design (Çepni, 2001). In Turkish education system, it is not generally possible or practical to control all the key factors and to select the students for the experimental and control groups randomly. Therefore, a quasi-experimental research design was preferred in this study since students were not randomly assigned to groups, and this study have already used intact groups of students. One of the two 11th grade classes in the same school was randomly assigned as an experimental group and the other as a control group. While the instruction in the experimental group was carried out by CSCL activities using GeoGebra software, no intervention was made to the control group and the instruction in the control group was taught by using textbook-based direct instruction, which was direct instruction as lecturing and answer–question mostly and using textbook-based activities. A pre-test and a post-test were applied to both groups before and after the experimental process.
Towards a Trait Model of Video Game Preferences
Published in International Journal of Human–Computer Interaction, 2018
Gustavo Fortes Tondello, Deltcho Valtchanov, Adrian Reetz, Rina R. Wehbe, Rita Orji, Lennart E. Nacke
In the first study, our goal was to devise a player traits model that is grounded on empirical evidence, is applicable to diverse types and genres of games, and is based out of survey items that are readily available to researchers. Additionally, we also intended to estimate how players score in these traits on average and how these scores are influenced by demographic factors, such as gender and age. Therefore, we conducted a cross-sectional correlational study based on the available data. A correlational study is a type of empirical research that aims to answer questions about the association between observed variables (Landers & Bauer, 2015). It differs from experimental or quasi-experimental studies, in which some variables are manipulated by the researchers to test a causal question, because there is no variable manipulation in a correlational study. Furthermore, a cross-sectional study design is employed in correlational research when the survey contains all the variables of interest.