8- NON-EXPERIMENTAL .pdf

NONEXPERIMENTAL RESEARCH DESIGN Dr. Latifah Alenezi Assistant Professor PURPOSE  Construct a picture of a phenomenon  Make account of events as they naturally occurs  Investigate current status of something  Describe existing phenomena without manipulating the condition Non-Experimental Re...

NONEXPERIMENTAL RESEARCH DESIGN Dr. Latifah Alenezi Assistant Professor PURPOSE  Construct a picture of a phenomenon  Make account of events as they naturally occurs  Investigate current status of something  Describe existing phenomena without manipulating the condition Non-Experimental Research Designs Descriptive Inter-Relational Studies Epidemiological Developmental Survey Descriptive Research Design  Univariant  Exploratory  Comparative Inter-Relational Research Design  Correlational  Predictive Developmental Research Design  Cross-Sectional  Longitudinal Epidemiological Research Design  Case Control  Cohort Survey Research Design  Comparative  Correlational  Explorative  Descriptive Descriptive Research Design Descriptive Research Design  Intend to examine and describe an issue.  Descriptive research design is a scientific method which involves observing and describing the behaviour of a subject without influencing it in any way.  Use to answer questions in ‘wh’ word: what occur, when occur, why occur, how occur PURPOSE  To observe, describe and document aspects of situation as it naturally occurs.  To serve as starting point for hypothesis generation or theory development.  Describe a Phenomenon CHARACTERISTICS  Especially effective when the area has been previously studied. Example- descriptive study of patients with shoulder dysfunction referred to physical therapy.  Data is collected through questionnaire, interview, files, document or observation. TYPICAL ANALYSIS  Descriptive statistics (graphs, charts, tables, Percentages, Frequencies etc.)  Textual analyses (content analysis) UNIVARIANT DESCRIPTIVE DESIGN It describe frequency of occurrence of a phenomenon rather study relationship It is not necessary that to study a single variable, but may be one or more variable This design mainly try to identify & describe perception, awareness, behaviour, attitude, knowledge and practice. Ex: PT Student awareness on BLS. Example A researcher is interested to assessing the experiences of patients suffering with rheumatoid arthritis. Frequency of different symptoms experiencing by the patients Type of treatment received during the course of disease, etc… There are multiple variables in this study. EXPLORATIVE  It identify, explore, and describe the existing phenomenon and related factors. Ex: Factors and morbidity pattern among LBW Babies  It simply describe the phenomena and indepth relate the cause of phenomena.  Aims to explain why particular phenomena work in the way that they do. EXPLORATIVE Exploratory-descriptive designs, usually field studies in natural settings, provide the least control over variables. The data collected either contribute to the development of theory or explain phenomena from the perspective of the persons being studied. COMPARATIVE  Compare and contrast two or more variable on same phenomena Ex: Perception of Labour Pain among Primipara & Multipara Mother  Attribute such as knowledge, practice, perception, attitude, symptoms, complication are compared Example A comparative study on health problems among older people in two selected different districts. Why Use Descriptive Research Design? • Define subject characteristics • Measure Data Trends • Comparison • Validate existing conditions • Conducted Overtime IMPORTANT ISSUES IN DESCRIPTIVE DESIGN  Cannot make conclusions about relationships studies. (snapshot of the current thoughts)  Subjects and instrumentation must be clearly identified.  Watch for graphic distortion.  Response and Non-Response Bias. INTER RELATIONAL STUDIES/RELATIONAL STUDIES INTER RELATIONAL /RELATIONAL STUDIES  It provides a deeper insight into the phenomenon of interest than descriptive studies.  It traces the interrelationships between variables.  A study that investigates the connection between two or more variables. INTER RELATIONAL /RELATIONAL STUDIES  Investigate the degree to which differences or variation in relation to two variables Ex : Relation between diabetes and development of hypertension  It determines degree of relationship.  It allows description and prediction of relationships.  Measures two Variables.  Examine the strength of relationship between variables CORRELATIONAL DESIGN  Describe a linear relationship  Do not suggest a cause-and-effect relationship  Do not suggest that variables share something in common Ex: Pill users and ovarian cancer  Indicates the extent to which one variable (x) related to another (y).  Magnitude and the direction of relationship are indicated by correlation coefficient. CORRELATIONAL DESIGN  It may be positive or negative ranging from (-1.00) perfect negative correlation to (1.00) perfect positive correlation and ‘0’ indicate no relation Ex: cervical proprioception and hand grip in physiotherapists with and without mechanical neck pain TYPES OF CORRELATIONAL RESEARCH DESIGN 1. Prospective research design 2. Retrospective research design 1. Prospective Research Design:  A design in which the researcher relates the present to the future is a prospective research design .  Prospective designs are often longitudinal but may also be cross sectional.  More reliable than retrospective b/c of the potential of control of data collection methods  Disadvantage: time and expenses of long periods of data collection 1. Prospective research design:  For example, researcher conducted a study to see the relationship between activity and the development of breast cancer in 1000 women from 1973 to 1993.  During which time they collected data on the incidence of breast cancer, activity level, medical history, education and other mental health factor to determine which factors were related. 2. Retrospective research design:  A design in which the researcher studies the current phenomenon by seeking information from past is a retrospective research design.  In this the researcher links the present phenomenon with the past events.  In other words, the researcher has a backward approach to study a phenomenon. 2. Retrospective research design:  For example, this design was used in ‘a retrospective correlational study on substance abuse-related high-risk factors among traumatic head injury patients admitted in neurosurgery ICU.  In this study, the researcher first approached head injury patients, then tried to identify the number of head injuries that occurred under the influence of substance abuse. CORRELATIONAL DESIGN ADVANTAGES  Analyse relationship among large number of variables  Demonstrate degree of relationship between variables  It opens up a great deal of further research to other scholars. PREDICTIVE STUDY  Aims to identify pre – existing characteristics (independent variable) that is predictive of a relationship with dependent variable. Ex: Intelligence level with performance, Obesity with hypertension  It uses retrospective data from one group to make prediction about similar group.  Shows how one variable predict score on another in time  Allow to calculate the value of one variable (criterion variable/dependent) based on value of another (predictor variable/independent) PREDICTIVE STUDY  They allow you to make estimates.  Length of prediction – shorter the time more the prediction  More time =more opportunity for another variable to influence. PREDICTIVE STUDY Typical Analysis includes: • Regression • Multiple Regression • Discriminant Analysis (whether heavy, medium and light users of soft drinks are different in terms of their consumption of frozen foods) DEVELOPMENTAL STUDIES DEVELOPMENTAL STUDIES  Involves the description of developmental change and the sequencing of behaviours in people over time.  It is generally used to adjunct research design with other designs, descriptive, cross sections, cross sectional survey, longitudinal co-relational study. DEVELOPMENTAL STUDIES  It is a study of status and is widely used in education, nutrition, epidemiology, and the behavioral sciences.  Problems can be solved, and practices improved through observation, analysis, and description. EX: study life span development have contributed to an understanding of psychological growth through old age. TYPES OF DEVELOPMENTAL DESIGN 1. Cross-sectional design 2. Longitudinal design CROSS SECTIONAL DESIGN  Examine data at one point in time.  Data is collected only on one occasion with same type of subject at one point or multiple points at one time Ex: Effect of aging on body fat distribution and CVR risk factors across age groups.  Data are compared in respect to socio- demographic.  Identify possible group differences.  But cohort effect (life experiences) can't be ruled out LONGITUDINAL DESIGN  It is used when data is required for several times at interval from same subject .  Ex: Physical growth of a child from birth to 18 years  The data is collected over an extended time period  Examine development by making a series of observations or measurements over time ADVANTAGES LONGITUDINAL DESIGN To study developmental changes /effect. DISADVANTAGES Time, cost , patience and sample turnover and other confounding variables EPIDEMIOLOGICAL STUDIES EPIDEMIOLOGICAL STUDIES  Study about distribution and causes of diseases in population.  Conducted to investigate causes of different diseases either by retrospective approach /case control studies, prospective approach / cohort studies  The cohort study is concerned with frequency of disease in exposed and non-exposed individuals, the case-control study is concerned with the frequency and amount of exposure in subjects with a specific disease (cases) and people without the disease (controls). EPIDEMIOLOGICAL STUDIES Cohort studies (Prospective studies):  In this design, a longitudinal approach is used to investigate the occurrence of a disease in existing presumed causes.  For example, a researcher longitudinally observes the smokers for development of lung cancer. COHORT STUDIES EPIDEMIOLOGICAL STUDIES Case-control studies (Retrospective studies):  In this design, causes of a disease are investigated after the occurrence of a disease (Researcher go back to the causes for a definite occurrence).  For example, a researcher investigates the history of smoking in patients diagnosed with lung cancer. SURVEY DESIGN /APPROACH  It is a popular method to describe and explore the events.  Used to collect information from a SURVEY DESIGN homogenous group of subject.  It ask subject to report their attitude, opinions, perceptions or behaviours  It gather current required data for obtaining new information. Types of Survey SURVEY DESIGN 1. Descriptive 2. Exploratory 3. Comparative Survey 4. Co relational Survey Ex: Cervical Cancer Describe - incidence , magnitude, characteristics Compare - Cx Cancer before and after age of 40 Exploratory – Cause of Cancer cx Co-relational – Multiple Sex partner and Cx Cancer  Data is collected by phone, mail, interview, questionnaire, etc.  For standardization of questionnaire must be validated. SURVEY DESIGN  Although mostly used in descriptive studies, can be used as part of both descriptive and predictive/relationship studies Two different types: • Cross-Sectional • Longitudinal CROSS SECTIONAL SURVEYS  Studies a phenomenon as it occurs at one point SURVEY DESIGN in time (e.g., political surveys, attitude surveys, etc.)  Studies a phenomenon through subjects that are along different timelines of the phenomenon (e.g., attitudes of managers in different age groups, surveyed at the same time) Longitudinal Surveys  The same group of subjects are studied over a period of time SURVEY DESIGN  Study of Morbidity, Mortality, Treatment Adherence etc.  In a longitudinal study, researchers repeatedly examine the same individuals to detect any changes that might occur over a period of time.  Longitudinal studies are a type of correlational research in which researchers observe and collect data on a number of variables without trying to influence those variables. ADVANTAGES OF NON-EXPERIMENTAL DESIGN  Require no experiment  Tend to be closest to real-life situation  Numerous human characteristics are inherently not subject to experimental manipulation (e.g. blood type, personality, health beliefs, medical diagnosis, etc.)  There are many variable that could technically be manipulated, but manipulated is forbidden on ethical grounds.  Easy and convenient to conduct.  Suitable to many field like psychology, sociology, education, nursing, etc. DISADVANTAGES OF NONEXPERIMENTAL RESEARCH DESIGN  Data are less reliable  The results obtained & the relationship between the dependent and independent variable can never be absolutely clear and error-free.  Nonexperimental studies are conducted for comparative purposes using nonrandomly selected groups, which may not be homogeneous and tend to be dissimilar in different traits or characteristics, which may affect the authenticity and generalizability of the study results.