Stats Week 2

Three essential characteristics of an experiment:

Three essential characteristics of an experiment:

1.Independent variable

2. A control group must be incorporated into the design

3.Subjects must be randomly assigned to groups

Variable

A characteristic that can be manipulated or observed and can take on different values (qualitative or quantitative)

Independent variables (aka explanatory variables)

The condition, intervention, or characteristic that will predict or cause a given outcome

Dependent variable

The variables that might change in response to some intervention. usually the result or outcome

Studies can have __ dependent variables, and ___ independent variables

multiple, multiple

Control group serves as

validation check for results

control groups are

an absolute requirement in an experiment

Random assignment

groups should be equivalent and made by chance

Bias

can have significant impact on results of study

Single blinding

only the researcher knows which treatment the participants received.

Double blinding

neither subject nor investigator know who received treatment or placebos

Make data and figures for your research ___ quality as you go

publication

Every journal has

its own rules for publication requirements

Why can't you submit your research article to multiple journals at the same time?

may cause conflicts in copyright

When submitting to journals, you shoudl

start by aiming high towards high impact journals, then move down

The abstract for a journal is usually written

last

Abstract

•Overview and purpose
•General description of methods
•Highlight of results
•Significance of results
•General conclusions

Introduction

•Statement of the problem
•Clinical relevance
•Review of literature
•Rationale and theoretical framework
•Specific purpose and hypothesis

Methods

•Study design
•Criteria for and methods of subject selection
•Description and number of subjects
•Reagents along with their volumes and concentrations
•Measurement methods and data collection techniques
•Data analysis procedures (stats)

Results

•Narrative description of statistical outcomes
•Tables and figures that summarize findings
•Statements to support or reject hypotheses

Discussion/Conclusions

•Interpretation of statistical outcomes
•Discussion of clinical significance
•Importance of the work
•Comparison of results with work of others
•How results support or conflict with hypothesis
•Critique study limitations and strengths
•Suggestions for further study

Acknowledgments

•Technical assistance
•Resources provided
•Funding source

References

•List of all references cited in the article

Tables should have

title and possibly footnotes

Figures and graphs should have

Legends which contain
•Title of figure
•Brief description of experiment
•Describe what is shown with labels or a key

Writing style rules

•Simplicity of language
•Don't be boring, but be concise
•Avoid "lab jargon"
•Define abbreviations

When is the use of first person acceptable in research?

Only when emphasizing own actions, experiences, assumptions, or opinions in the discussion/conclusion

(don't use I unless you are the only author)

What is the purpose of a Literature Review?

to concisely perform a systematic review of the current progress in the field

Literature reviews must use information from ___ sources

recent

Literature reviews provide

an argument or interpretation of the "state of the field"

Review article format usually includes

Introduction and background

Body of the paper

Conclusion (summarize main points)

Acknowledgements if applicable

References (usually a lot, >20)

Poster presentations usually represent a research projects

initial findings

In order to conduct a research project, one must

evaluate research reports
ie. serve as a peer review for grants or papers

As a peer reviewer, you must evaluate (7)

1.Importance of research
2.Originality
3.Appropriate research design
4.Adequacy of methods
5.Soundness of conclusions and interpretation
6.Relevance of discussion
7.Clarity in writing

To find out if a research article is relavent, you should read the ____ and ____ first

title, abstract

A research design and goals are often determined by

–Availability of subjects

–Availability of samples

–Ethical concerns

–Cost

Case studies

description of new, interesting, and unique clinical cases as a means of sharing special information to professional colleagues

Developmental study

description of developmental change and sequencing of behaviors over time

Longitudinal study

continuous, repeated measurements following particular subjects over a prolonged period of time

Cross-sectional study

observations of different age or developmental groups at one point in time. Makes inferences

the two groups in an exploratory research design are

–The group displaying an aspect to be studied

–Control/comparison group

Case-Control Study

groups are selected on the basis of whether or not they have the disorder being studied

In a case control study the groups are

–Cases have the disorder

–Controls are the comparison group without disorder

Case-Control Study purpose

–Purpose is to determine why cases have the disease and controls do not

Cohort study

two groups are chosen, one which has been exposed to a clinical intervention or agent, by chance or choice

Cohort study purpose

–Establish relationship between exposure and risk of developing a particular disease

Randomized controlled trial (RCT)

is a true experimental design in which subjects are randomly assigned to at least two comparison groups

Randomized controlled trial (RCT) groups are

Experimental group - intervention applied here

Control group

What is the strongest research design?

Experimental research