Naslov
Experimental Research

Autori
Mildner, Vesna

Vrsta, podvrsta i kategorija rada
Poglavlja u knjigama, znanstveni

Knjiga
The SAGE Encyclopedia of Human Communication Sciences and Disorders

Urednik/ci
Ball, Martin J. ; Damico, Jack S.

Izdavač
SAGE Publishing

Grad
Thousand Oaks (CA)

Godina
2019

Raspon stranica
728-732

ISBN
9781483380834

Ključne riječi
Ethics in Communication Disorders Research ; Qualitative Research ; Quantitative Research ; Significance ; Statistics: Descriptive ; Statistics: Predictive

Sažetak
Research methods are many and varied, and ideally, the choice is guided primarily by the nature of the problem to be studied. With respect to participants, research can be carried out on individuals (single-case studies) or groups. The latter may involve a single group or multiple groups (at least two: experimental and control). They can be further classified as between-subject/group or within-subject/group studies. With respect to type of data, there are quantitative (numerical data), qualitative (textual data), and mixed methods, which may be applied to nominal, ordinal, interval, ratio, or other type of scales. Number of variables will determine whether the study uses single-variable (one independent variable) or factorial (at least two independent variables) design. Based on the criterion of amount of control the design is categorized as nonexperimental (also referred to as pre-experimental), quasi-experimental, or true experimental. Nonexperimental studies allow the least amount of control because they deal with phenomena occurring in natural environments. Most of them are correlational: They try to determine correlations among relevant variables without manipulating them. With respect to phenomena studied, they can be classified into descriptive/exploratory survey studies and interrelationship/difference studies. Quasi-experimental design (most frequently used in applied research) allows more control over independent variable(s) and involves manipulation but lacks randomness in subject selection. (True) experimental design is characterized by manipulation of independent variable(s) and a high amount of control over subject selection and assignment.

Izvorni jezik
Engleski

Znanstvena područja
Filologija, Interdisciplinarne humanističke znanosti

POVEZANOST RADA