based on observations

why, whom, how, when

strive to avoid misleading results and poor interpretations

researchers give treatments

observe if they cause changes

cause and effect

Experimental Group

vs

Control Group

experimental- recieves treatment

control- does not recieve treatment

True Experiment

researchers do not give treatments

observe participants

most common type is a survey/poll

Demographics

background characteristics

Causal-Comparative study

nonexperimental

used when not possible to conduct an experiment

sometimes called ex post facto study

do not give treatments

look to the past for possible causes of a current condition

Types of Nonexperimental Research

qualitative and quantitative

causal-comparative

survey/poll

census

case study

longitudinal

correlational

historical

Survey/poll

describe attitudes, beliefs, and behaviors of a population

 Case study

intensive one on one study of individuals

Longitudinal research

repeatedly measure traits of the participants over a period of time in order to trace developmental trends

correlational research

degree of relationship among two or more quantitative variables

Variable

a trait or characteristic with two or more categories

mutually exclusive categories

exhaustive

a variable must have a category for each respondents opinion

independent variable

presumed cause, predictor

stimulus or input

independent variable causes changes in the dependent variable

(some researchers refer to any variable that comes 1st as independent)

Dependent Variable

criterion- "standard"

response or outcome

independent causes changes to dependent

some researchers refer to any variable that comes later than the 1st variable as dependent

Variables in Experimental Studies

all experimental studies have at least one independent and one dependent varible

researchers physically manipulate independent variables

(in nonexperimental researchers do not physically manipulate, they observe)

purpose of experiments

to estimate the extent to which independent variables cause changes in dependent variables

physically manipulate

physically administer treatments

research hypothesis

a prediction of the outcome of a study

simple research hypothesis

predicts a relationship between two variables

directional hypothesis

researchers predict which group will be higher or have more of some attribute

nondirectional hypothesis

there will be a difference but does not predict the direction of the difference

What can be used instead of a nondirectional hypothesis?

research purpose

research question

(all equally acceptable in scientific community) 

Conceptual definitions of variables

dictionary definition

operational definition

when a researcher operationalizes a variable

operationalizing- redefining a variable in terms of physical steps

(if a researcher fails to provide operational definitions of all key variables, the research report is vague at best and results are limited in usefulness)

replication

an attempt to confirm the results of a study by conducting it again in the same way

(researchers try to provide operational definitons that are adequate to permit a replication)

when to use quantitative research

economic variables

potential participants not avalible for extensive interactions/observations

time/funds limited

quick snapshot of a problem

when to use qualitative research

emotional impact

little known about topic

participants belong to closed/secretive culture

What does “evidence-based” mean??

Evidence-based Practice is based on 3 things:

1. The best available scientific evidence

2. A dental professional’s clinical skill and judgment

3. The patient’s needs and preferences

the evidence pyramid

(in order of top to bottom)

Systematic reviews and Meta-Analyses

randomized controlled double blind studies (RCT)

cohort studies

case control studies

case series

case reports

ideas, editorials, opinions

animal research

in vitro (test tube) research

Systematic reviews

A comprehensive survey of a topic in which all primary studies of the highest level of evidence have been systematically identified, evaluated, and then summarized according to an explicit and reproducible methodology

Meta-analyses

A survey in which the results of similar studies are combined statistically and analyzed as if they were one study.

RCT

Two groups: 1 treatment and 1 control

Subjects are randomly assigned to each group

Assigning subjects at random reduces the risk of bias and increases the probability that differences between the groups can be attributed to the treatment.

(RCTs are the standard method of answering questions about the effectiveness of different therapies.  If you have a therapy question, first look for a RCT, and only go on to look for other types of studies if you can’t find one)

Quasi-experimental Studies

Patients currently have a certain condition and/or are receiving a particular treatment

Followed over time

Compared to another group who are not affected by the condition under investigation.

Not as reliable as RCTs

Causal-comparative Studies

Subjects already have a certain condition

Are compared to people who do not have the condition

Less reliable than RCTs and Quasi-experimental studies

Sources of Evidence

Primary

Firsthand testimony or direct evidence of a topic under investigation

Example: published articles found in databases (i.e.: Medline)

Secondary sources of evidence

Summaries and analyses of evidence derived from and based on primary sources

Evaluates, interprets, and analyzes primary data

One step removed from the evidence

Example: systematic review, meta-analysis

Parts of a published research report

Title

Names of researchers/ Authors w credentials

Abstract

Keywords

Background (not always present)

Body

Acknowledgment

References

Body of a Published
Research Report

introduction

Literature Review(sometimes included in intro)

Materials & Methods

Results

Discussion  

Summary

Conclusions (sometimes included in summary)