Internal Validity

• exists if oberved effects of independent variable on dependent variable are real and not caused by extrneous factors

• Threatened when there are alternative explanations for effect of independent variable on dependent variable

External Validity

• concerned with ability to gerneralize study results to other groups and settings beyond those in current experiment

• research that can be generalized will typically have more practical value to the professional than research that can't be generalized

Threats to Internal Validity

• extraneous factors that allow for alternative explanations as to waht caused a given efect on dependent variable
• various strategies can be utilized to control threats

History Effect

• occurs when events between pretest and postest of a research study could affect particpants in such a way to impact dependent variable

• examples that could affect participants include stress, natural disasters, news events, illness

Groups

• Control group- group of particpants exactly like treatment group excpet they recieve no treatment

• Experimental group- group recieving the treatment

Maturation Effect

• occurs when changes are seen in subjects due to time elapse since beginning of study; may not be result of any program effects

• over time particpants may have become older, wiser, stronger, or more experienced

Testing

• Baseline- measurement taken to obtain the status or level of a variable prior to initiating a study

• Test-wise- when particpants have learned how to do better on a test

• Testing effect- differences seen from baseline to posttest

Instrumentation

• can cause threats to internal validity when measurements are not accurate or procedures aren't standarized

• well-designed intruments reduce threat

Selection Bias

• reduces internal validity due to individuals being selected in nonrandom manner; participants in both groups may have differences before start of study

• randomly assigning participants to groups will help

Selection Maturation Effect

• combines selection bias with maturation effect; occurs when intact groups with varied maturation levels are used

• one group may be more mature, intellectual, etc.

• pretesting and prescreening groups on maturity levels can help

Statistical Regression

• occurs when particpants are selected on basis of extremely high or low scores

• those who score highest have nowhere to move but down; those who score lowest have nowhere to move but up

• to counter, follow random samples rather than high or low groups

Mortality

• illness, death and dropout rates
• difficult to counter
• obtain good demographic information at beginning of study
• after study determine if group makeup changed after mortaliyy occured

Hawthorne Effect

• occurs when particpants' attitudes toward being involved in a study affect they way they behave

• to counter, keep participants from knowing they are taking part in study or being observed

Placebo Effect

• caused by particpants' expectations rather than by provided treatment

• placebo is an inactive substance

• blind study- particpant doesn't know if taking real drug or placebo

• double blind study-resarcher doesn't know who is taking drug or placebo

Diffusion Effect

• occurs when treatment being applied to one group spills over or contaminates another

• be careful in selecting groups and group interactions

Location Effect

• occurs when there are differences in locations where interventions take place

• best method for control is to make location the same for all participants

Implementation Effect

• involves individual(s) responsible for implementation of study; they may inadvertantly introduce inequality or bias

• if multiple people they may not be equal in knowledge, training, understanding, personality, etc.

• if individual(s) shows favors one group over other

Threats to External Validity

Selection, Setting, and History

Treatment Interactions

• Selection- ability of researcher to generalize results of study beyong groups involved

study population- group with random sample population- whole group

• Setting- extent to which environmental conditions can be duplicated in other settings

• History- researcher tries to generalize findings to past and future situations

Three Strategies to Control Threats to Internal Validity

• randomly select participants fromm well-defined study population

• randomly assign selected participants to groups

• include non-treatment control groups in research design

External Validity

• carefully consider what groups one can generalize to; do not generalize to oustide groups

• always duplicate setting and historical factors as accurately as possible when replicating progam with different population

External Validity cannot exist without Internal Validity

• researchers should always be concerned about ensuring internal validity before trying to establish external validity