Research Exam 3

Methodology

Design and variables

Research approval process

Participants

Equipment

Materials

Procedures

Data collection and analysis

Necessary appendixes

Developing a purpose statement

Guide to your research

Must be developed based on the literature that has been reviewed

(Hypothesis, statement of purpose, purpose statement, research question(s))

Three basic ethical principles of Belmont Report

Respect for persons

Beneficence

Justice

When was the Belmont Report published?

April 18, 1979

Who was the Belmont Report written by?

The National Commission for the Protection of Human Subjects of Biomedical and Behavioral Research

How many subparts are there?

4

Subpart A

Common rule

Protections for research subjects

Subpart B

Additional protections for research with pregnant women and fetuses

Subpart C

Additional protections for research with children

Subpart D

Additional protections for research with prisoners

Respect for persons

Acknowledge and treat individuals as autonomous

Protect those with limited autonomy

How is respect for persons ensured?

Informed consent

Beneficence

Do not harm

Benefits should outweigh the risks

Justice

Treat people fairly

Design research to share burdens and benefits equitably

randomization

Research

A systematic investigation, including development, testing and evaluation, designed to develop or contribute to generalizable knowledge

Human subject

A living individual about whom an investigator conducting research obtains

Human subject research

1) information or biospecimens through interventions or interactions with the individual, and uses, studies, or analyzes the information or biospecimens, or 2) obtains, uses, studies, analyzes, or generates identifiable private information or identifiable biospecimens.

Minimum risk

Probability and magnitude of harm or discomfort anticipated in the research are not greater in and of themselves than those ordinarily encountered in daily life or during the performance of routine physical or psychological examinations or tests.

Informed consent

Subject is informed in writing of the procedures and expectations during the project

Generalizable knowledge

Information where the intended use of the research findings can be applied to populations or situations beyond that studied

ASHA Implications

Maintain research records

Permit voluntary participant participation

Ensure proper working of equipment used in research

Not misrepresent research results

Not misrepresent contributions to research

No relationships with research participants over whom they have professional authority or power

•Conflict of interest - bias

Fabrication

Making up data or results and recording or reporting them - making up things that does not exist

Falsification

manipulating research materials, equipment, or processes, or changing or omitting data or results such that the research is not accurately represented in the research record

EX: deleting outliers

Plagiarism

the appropriation of another person's ideas, processes, results, or words without giving appropriate credit

IRB

Institutional Review Board

Protects subjects in any type of research

IRB membership

At least 5 members

Members from varied backgrounds and professions

Non-scientist member

Scientist member

Unaffiliated member

Federal regulations stipulate that an IRB can:

Approve research

Required modifications (to secure approval)

Disapprove research

Conduct continuing reviews consistent with federal requirements and institutional policy

Verify no material changes occurred since previous review

Observe, or have a third party observe, the consent process and research

Suspend or terminate approval

Types of IRB Review

Limited IRB Review of Select Exemption Categories

Expedited Review

Convened Committee Review

Limited IRB Review of Select Exemption Categories

The research may not involve more than minimal risk.

The entire research project must be consistent with one or more of the federally defined categories (OHRP 2003).

Expedited Review

Used for the published categories and also for minor changes in previously approved research during the period for which approval is authorized (for example, an amendment), and for limited IRB review of select exemption categories.

Convened Committee Review

The initial review of all studies that are not eligible for expedited review or exemption status.

Basic elements involved in Informed Consent

Statement of purpose

Brief description of procedures

Description of any reasonably foreseeable discomforts or risks

Statement regarding compensation if injury occurs.

Description of benefits for subjects and others

Disclosure of alternative procedures

Offer to answer questions/inquiries

Statement that participation is voluntary

Participation may be withdrawn at any time without any penalty or loss of benefits.

Statement of confidentiality

Informed consent areas

Name of principal investigator(s)

Sponsoring departments and phone numbers

Name of person giving consent

Signature of subject (If child, it goes to parent)

Children and assent for study

Provide subjects with a copy of the form

Indication of commitment of time

Must be obtained before any research is started.

Two aims of research design

Classifying behaviors in a given research area

Identify the variables involved in explaining the behavior of interest

Variables in Research

Independent

Dependent

Extraneous

Control

Independent variabeles

Presumed cause, what you are manipulating

Researcher controls this variable

The control/selection may impact the DV

Dependent variable

Behavior that is changed; measured

Researcher can not control this variable

Extraneous variables

Intervening factors may impact the study

Control variables

Used to control the extraneous variables

Research strategy

General plan

Have this before a design

Research design

The specific tactics you use to carry out the strategy

Levels of EBP

4

Strong/Strongest

Moderate

Limited

Weak

Strongest

Meta-analysis of multiple well-designed controlled studies

Strong

Well-designed randomized controlled trials

Moderate

Well-designed non-randomized controlled trial (quasi-experiments)

One area of true experiment missing (usually control grou)

Limited

Observational studies with controls (retrospective studies, interrupted time-series, case-control studies, cohort studies with controls)

Missing two parts - usually only has control group

Weak

Observational studies without controls (cohort studies without controls and case series)

No criteria of true experiment

Characteristics of Research Design

Counterbalance

Longitudinal

Replication (reliability)

Duplication

Developmental

Hawthorne Effect

Counterbalance

A procedure that allows a researcher to control the effects of nuisance variables in designs where the same participants are repeatedly subjected to conditions, treatments, or stimuli

EG: within-subjects or repeated-measures designs

Need to do this to those factors that may influence our study results

Need to make sure what we measure is due to our treatment and not something else

Longitudinal

Observational study

Once started, the researchers do not interfere with the subjects. The researchers could conduct several observations of the SAME subjects over a period of time. Could be many years

Principle of triangulation

Replication (Reliability)

The deliberate or conscious repetition of research efforts, intended to confirm or extend previously or simultaneously obtained, but still uncertain, findings

Not only legitimate but essential, providing "Proof positive" of otherwise uncertain research findings and lending confidence to conclusions drawn from them

Testify the findings of being correct

Duplication

Inadvertent, unconscious, or more rarely, deliberate repetition of research efforts, thus not serving a need to confirm or otherwise verify conclusions from previous research undertakings

Inherently wasteful and is normally guarded against, although it can also be excused or even encouraged under special circumstances

Developmental

Evaluating changes in behavior that relate to changes in a person's chronological age

Particularly true for children because they change so fast

Hawthorne Effect

Participants change their response behaviors simply because they know what is to be studied

Strategies that participants develop during the experiment

Change stimuli!!

Non-experimental designs

Descriptive

Comparative

Observational

Case Study

Developmental

Archival/Historical (Retrospective)

Content Analysis

Creative

Survey

Cover Letter

Qualitative

Correlational

Experimental Designs

Between-subjects

Within-subjects

Single-subject

Descriptive desings

Used to describe an event, condition, or relationship

Could be combined in other methods

Must provide an accurate description

Researcher must explain the concept, not just give it a name

Comparative design

Used to measure the behavior of two or more types of participants

Looks at similarities and differences - purpose is to find out these similarities and differences

Observing phenomenon and then building up the relationship

Combine with further analysis

Observational research

Describe the behaviors being observed

Develop a recording sheet

Establish reliability of observations by training observers

Not true experiment

Within observational research, you may observe:

Frequency of behavior

Duration of behavior

Interval between behaviors

Within observational research, you may incorporate:

Time sampling: specific period of time

Individual sampling: select individual person

Event sampling: only one behavior

Case Study design

Intended for a N of one or a very small group

Identify specific goals and objectives

Brief overview of literature regarding the topic

Detailed methodology to describe program of intervention

Results will describe the data collected

Charts may be used

Discussion presents the current status of the case after the intervention

Conclusions would include further goals and objectives for the case

Looking at interaction of behavior, not a true experiment

Developmental design

Measure changes in behavior or characteristics over a specified period of time

Could be quasi-experimental design sometimes (usually randomization and manipulation)

May be influenced by age

Maturational changes

3 types

3 types of developmental design

Cross sectional

Longitudinal

Semilongitudinal

Cross sectional

EG: select participants from each of a number of age groups

Made as subject ages and look at differences as you go

Longitudinal

A single group of participants if followed over some time period

Semilongitudinal

Cohort-sequential

Features of both the cross-sectional and longitudinal

Archival/Historical

Uses history as the data base

Involves a great deal of library research

Develop a guide for review

Must systematically review the literature

Each source is reviewed in same manner

Results is the largest section

Content analysis

Used to analyze written or spoken record for the occurrence of specific behaviors/categories of events

Must define characteristics (objective with clear rules, systematic, generality)

Operationally define concepts

Useful for understanding behaviors

Creative Research

Completed when a product is in the end result

Incorporates a creative process

Can be time consuming

Incubation phase

Illumination phase

Survey Research

Conducted on people who have information you want

Evaluates the behavior tendency and attitudes of your participants

Minimum of 50 participants

Have a researcher pre-test it

Survey research may be done via:

Mail/email

Interview

Internet (web-based)

Telephone

Survey research must have clearly designed questions:

Clearly define the purpose of the study

Keep the questions focused to the purpose

Arrange neatly and print dark

Obtain demographics

Open- and close- ended

When choosing a survey, think about:

1. Is it appropriate

2. Adequacy of sample surveyed

3. Threat of people who don't return survey

4. Good questionnaire

Cover letter components

Introduction

Sponsorship

Purpose

Criteria for selection

Use of data

Confidentiality

Appeal for cooperation

Include a return data or a completion date

Express gratitiude

Response rates

Mailed surveys

Post card reminders

Post card surveys

Internet surveys with email reminders

Why people respond to surveys

Recognition

Altruism

New experience

Catharsis

Extrinsic rewards

SASE

Qualitative research method

Involves subjective observations of natural situations that are analyzed and interpreted for meaning

Permits triangulation of data

No statistical analysis

Series of data collection

Each session has questions developed from previous session

Caution of subjectivity

Triangulation

Refers to the use of multiple methods or data sources in qualitative research to develop a comprehensive understanding

Helps avoid subjectivity

In qualitative research, data may be gathered through:

Observing

Videotaping

Interviews

In qualitative research, you may be looking for:

Ethnographic data

Phenomenological data

Ground theory

Hermeneutics

Ethnographic method

Discover categories, subcategories, components of meanings, may cover themes

Phenomenological methods

Individual experiences such as ange, imagination, approval

Read through informants statements and get sense of the whole

Grounded theory

Use the ethnological and phenomenological data and build a theory

Hermeneutics

Determine overlapping areas and see what is interconnected

Correlational design

Studies relationships among two or more variables

Change in one variable can be predicted from another

Data are analyzed with correlational statistics

No cause/effect relationship

Between-subjects design

Each treatment is administered to a different group of subjects

Each group you have different participants

Within-subjects design

A single group of subjects is exposed to all of the treatments, one treatment at a time

Same group of participants receiving all the treatment

EX: pre/post treatment

Participant behavior is measured repeatedly, also called repeated-measures design

Types of within-subjects design

Single-factor two-level design

Single-factor multilevel design

Single-subject design

Similar to the within-subject design in that subjects are exposed to all treatments

Do not average data across subjects - no statistics used - we just use visual inspection/observations

Used with less than 18

Focus is on the analysis of performance by an individual

Repeated measures designs

Baseline (A) - Intervention (B) - Withdrawal (A_

A-B-A-B-A

Between-subjects designs

Single-factor randomized groups design

Randomized two-group design

Randomized multigroup design

Matched-groups design

Single-factor randomized-groups design

Assign your subjects randomly to different groups, then expose each group to a different level of IV while otherwise treating them as alike as possible

EX: treatment vs no treatment

Randomized two-group design

Randomly assign subjects to two groups, expose the two groups to different levels of the independent variable, and take steps to hold extraneous variables constant

Randomized multigroup design

Add one or more levels of the independent variable

Matched-groups design

Matched sets of subjects are distributed at random, one subject per group, into the groups of the experiment

The requirement for pretesting and matching makes the matched design more demanding and time consuming that the randomized design

Combination of randomized groups design and matched-groups design:

Matched-pairs design

Matched-multigroup designs

Cautions of within-subjects design

Carryover effects

Carryover effects

Occur when a treatment alters the behavior observed in a subsequent treatment

They develop strategies and change their normal response

Sources of Carryover

Learning

Fatigue

Habituation

Sensitization

Contrast

Adaptation

How to deal with carryover effects

Counterbalancing

Taking steps to minimize carryover

Making treatment order an independent variable