Ab Psyc: Chapter 3: RESEARCH 2/21/2013 11:09:00 AM
Research = systematic way of finding answers to questions
A Scientific Approach
Organized ways, backed up by data
Ppl working with children tend to have biased opinions on their own
beliefs which might lead to interpreting info incorrectly.
Simple explanations may appeal to parents/teachers cuz it promises
a quick fix for a complex problem.
Good reasons exist for skepticism within ab child psyc
o Experts on it tend to disagree with one another.
o Research studies are oversimplified, make things more
o Research findings are often in conflict with each other (diff
results for e.g.)
o Diff reccomendations on HOW children should be helped.
o Even when scientific evidence is valid, ppl may dismiss it cuz
of personal experience (e.g. Well my dad used his belt on me
and I turned out fine!)
Important: It is the accumulation of findings, not one study, that advances
When Science is Ignored
When science is ignored or dismissed, there are things like
facilitated communication that although controversial can help…
e.g. with holding child’s hand over keys of computer.. doesn’t
influence what key to press but rather helps them press it.
THIS IS A BAD THING (FC)… cuz it showed that when
listening to a diff message (the autistic child and the helper)
the answer was relative to the helpers message, not the one
given to the ear of the child!
It is a pseudoscience.. very unprofessional cuz it doesn’t help the child
improve and dismisses scientific theory… The diff b/t scientific and
pseudoscientific claims are not based on evidence or not but RATHER the QUALITY of the evidence, how it was obtained, etc. Science admits when
wrong and is open to change. Pseudosciences do not follow this.
The Research Process
• Begins with a hypothesis (based on a theory, observation, etc)
• Identifying sample, measurement ethods and design.
• Gathering and analyzing the data and interpreting results
Researchers must include research designs and methods of data analysis
that can identify direct and indirect effects and diff causal pathways for
Research questions and topics are often based on the theories of atypical
development and behavior.
When little or no theories are known, researchers may also develop a
question without an explicit prediction, e.g. “Are more children depressed
these days than a generation ago?”
Nature and Distribution of Childhood Disorders
Epidemiological research: the study of the incidence, prevalence, and co-
occurrence of disorders and competencies clinic- referred and community
Incidence rates : extent to which new cases of a disorder appear over a
specified period. (e.g. number of youths who develop a depressive disorder
during school year)
Prevalence rates: all cases, whether new or pre-existing , observed during
a specified period of time ( e.g. the number of teens with conduct disorder in
the general population during 2007 and 2008). Lifetime prevalence = whether children in the sample have had the disorder
at any time in their lives.
10-20% of all children have a disorder.
Overall rates obscure the enormous variability in reported rates from study
E.g. rates of reported problems in children by teachers are between
6-20% from teachers and 10-40% when reported by parents.
Prevalence estimates vary depending on what definition of some case (e.g.
problematic children) is used.
Estimates based on single symptoms being much higher than those
on patterns of symptoms.
Prevalence rates also vary depending if cases defined in terms of patterns of
symptoms or impairment in functioning, or both.
Fewer cases are identified when both are used than with either or.
If we don’t take (social economic status) SES into account we might
conclude that differences in learning are related to ethnicity when instead
they are from factors associated with an SES variable such as nutrition poor
or fewer learning opportunities.
E.g. African americans that are middle class have same rates as
caucasians middle class.
Three variables in ab child psyc are CORRELATES, RISK/PROTECTIVE
FACTORS, and CAUSES of other variables.
Most research in ab child psyc is designed to answer questions
about the relation between the three general variables and
Correlates are variables assoc. with a particular pt in time with no clear
proof that one precedes the other. A risk factor is a variable that precedes an outcome of interest and increases
the chances of a negative outcome.
A protective factor is a positive variable that precedes an outcome of interest
and decreases the chances that a negative outcome will occur.
Moderators are independent on relationship b/t two variables, whereas
mediators account for some/all of the relationship.
Moderator variables: influence the direct or strength of the relationship.
Examples: SES, child’s age, sex, cultural background
o Example with sex – Girls are more prone to internalizing
depression than boys.. therefore the relationship depended on
if adolescent was a girl or boy.
Mediator variables: the process, mechanism, or means thru which a
variable produces a particular outcome.
Mediators describe what happens at the psychological or
neurobiological level to explain how one variable results from
o E.g. In a study, Snyder found that on days when mothers
experienced negative moods and frequent hassles, they were
most likely to respond negatively to their children’s
misbehavior and to reinforce (unintentionally encourage) their
children’s coercive tactics during mother-child conflicts. The
use of this maternal discipline was related to an increase in
same-day child behavior problems.
Therefore, the relationship b/t maternal distress and
child conduct problems is partly mediated by the
disciplinary strategies mothers use on days they feel
Can be long-lasting outcomes for some that suffer at a young age. Interventions
Treatment outcomes are evaluated using randomized controlled trials
(RCT) – assigned diff treatments, equal chance of selection…
Treatment efficacy: whether the treatment can produce changes under
Treatment effectiveness: refers to whether the treatment can be shown
to work in clinical practice, not just in well-controlled research settings.
Clients usually referred than selected…therapists provide services
without many of the controls used in research.
Important: Treatment efficacy (wellcontrolled conditions) works better for
children than does TEffectiveness (clinical)
Methods of Studying Behavior
Standardization, Reliability, and Validity
Standardization: a process that specifies a set of standards or norms for a
method of measurement that are to be used consistently across different
assessments of the construct of interest (everything is conducted the same
way every time)
One type of reliability, internal consistency refers to whether all parts of a
method of measurement contribute in a meaningful way to the info
Interrated reliability – various ppl agree on what they see
Test-retest reliability – the results need to be stable over time
Validity is not all or none but rather a matter of degree, and can be assessed
in many ways. First way is its face validity – the extent to which it appears to
access the construct of interest.
o e.g. face valid if a questionnaire asks if you get nervous
before writing an exam (if of course it measures test anxiety)
as opposed to asking if you are a parrot.
Construct validity: whether scores on a measure behave as predicted by
theory or past research. E.g. low scores show low cases of something, high
scores show high cases of something.
Construct validity has two components: convergent validity and discriminant
Convergent validity reflects the correlation between measures that are
expected to be related – e.g. a teens report of her depression in interview
and her scores on a depression questionnaire.
Discriminant validity refers to the degree of correlation between measures
that are not expected to be related to one another.
e.g. scores on a measure designed to assess depression and
another designed to assess intelligence should not correlate.
Criterion-related validity refers to how well a measure predicts behavior in
settings where we would expect it to do so – at the same time (concurrent
validity) or in the future (predictive validity).
E.g. high scores on this means has anxiety and might have trouble
making friends in future,
CR validity tells whether scores on a measure can be used for their intended
Includes interviews, questionnaires, checklists and rating scales,
psychophysiological recordings, brain imaging, performance measures, and
direct observations of behavior. Reporting
Assess the perceptions, thoughts, behaviors, feelings, and past experiences
of the child, parents, and teachers.
Include unstructured clinical interviews, highly structured diagnostic
interviews, and quesitonnaires.
• self report measure
• informant-report measure
limitation – trouble for children who haven’t developed verbal skills
MUST be sensitive to the language and cultural background of the person
Show relationship b/t physiological processes and behavior to identify which
nervous system structures and processes contribute to children’s atypical
development and behavior.
Common measures: ANS activity like heart rate, blood pressure,
breathing, pupil dilation, etc.
Limitation – c