Studied by 29 peopleWhat are the ways we get knowledge according to Pierce?
something we've always known; cling to beliefs in the face of conflicting facts
appealing to authority; accepting knowledge bc they are authority
using intuition; just because it seems the right answer
science and empericism
What is the definition of scientific research?
it is systematic: a process of ruling out other explanations
it is controlled: outside influences limited or eliminated
it is empirical: direct observation or measurement
it is critical: sources of error are explained or built into research design
it is amoral: empericism isn't good or bad; it is valid and reliable or invalid and unreliable
it is public: accessible to other members of the scientic community for evaluation and repetition
What are the steps of the scientific method?
state the problem
form initial hypothesis and apply knowledge
form final research question/hypothesis
observe - test (experiment)
apply results
accept or reject
What is the difference between research question vs hypothesis?
Hypothesis is a statement, research question is stated in form of question
What are key components of a hypothesis?
state relationship between two variables, unambiguous, state how you will measure or observe the variables
What are key components of writing the research question?
needs to be answerable by observation and specify the how, explain the meaning of concepts- they should apparent or operationally defined, devise alternative explanations, reflect a controlled experiment
How is scientific research accomplished?
manipulating conditions and variables, contolling events, and replicating the results
Compare and contrast extraneous and confounding variables.
An extraneous variable is any variable that you're not investigating that can potentially affect the dependent variable of your research study. A confounding variable is a type of extraneous variable that not only affects the dependent variable, but is also related to the independent variable.
What is not an advantage of multivalent design?
The IV can be tested multiple different ways
Is more economical
Gives a broader picture when the IV is continous
The range of values for the IV is sampled more completely
is more economical
What is not an advantage of parametric?
It easily deals with interaction of multiple IVs on DV
It's more economical
It closely resembles the real world
The participant would spend more time being tested
The participant would spend more time being tested
What is not a disadvantage of bivalent designs?
They are limited in scope
May lead to erroneous conclusions because they don't encompass all potential effects
Results can't show curve functions
None of these
None of these
What is multivalent design?
More than two values of the independent variable is measured on the dependent variable; where there is multiple instances of the "same type" of IV (multiple therapy sessions, different durations of testing)
What is parametric research design?
Studying the simutaneous effects of multiple IV you predict will effect the outcome on one dependent variable; you can study the main effect (effect of each IV) and interaction of IV (combined effect of IVs toward each other)
What type of design describes how subjects are compared in analysis in a parametric research study?
Mixed: it's a combination of between subjects and within subjects
Contrast between subject design and within subject design.
Between subjects: 2+ separate groups are compared for how the independent variable affects the dependent variable (there is an experimental and control) Within subjects: compare the effect of the IV with different conditions on the same subjects
Give a research question that fits this design description: bivalent, within subjects, descriptive, A-B
What are teachers' in high schools' perceptions of speech language pathologists before and after a intraprofessional workshop developed by ASHA and the NEA, measured by a survey before and after?
Dependent variable: teachers' perceptions IV: intraprofessional workshop Bivalent: with and without the IV (the workshop) A-B: there is an observation of behaviors before and after the IV Descriptive: observing the relationship between variables (the workshop and perceptions) through a survey
Give a research question that fits this design description: parametric, between subjects, experimental, B
What is the interaction of intonantion therapy and 'memantine' in patients with aphasia compared to patients who recieve only intonation therapy and no therapy after 3 months? Parametric: the interaction of intonation therapy and memantine (main effect of each IV) IV: intonation therapy and medication, intonation therapy only, and no therapy Experimental: observes the relationship between intonation therapy and medication B: only observing/measuring with after tx condition (there is no initial eval) Between subjects: stroke patients with aphasia divided into 3 groups
Give an example of A-B-A time sequence design.
A-B-A is when behaviors are described before, during or immediately after tx, and then after a short period of no tx. An example of this would be measuring the effect of intonation therapy on a person with aphasia. A: gathering baseline data before administering the therapy, B: measure outcomes, A: reassess after a period of no treatment- was the therapy generalized/not generalized?
Historical data is ____ and descriptive and experimental data is _____
post-hoc apriori
There are two "versions" of within subject designs:
Single subject and multisubject
What is content validity?
Does the test accurately capture all aspects of the subject you want to study in order to assess it well? Does it sample a subjects ability in all areas of a certain schema or domain?
What is a construct?
A group of interrelated variables you want to study ie construct of aggression encompasses relational, emotional, physical aggression etc
What is construct validity?
The degree to which your test actually measures what it says it does.
What is concurrect validity?
does it agree with established/normed measures?
What is external reliability?
does it apply to a larger population
Give an example of intrasubject reliability
A subject who takes the GFTA twice gets a standard score of 63 the first time and 64 the second time
Give an example of intrajudge reliability.
a gymnastics competition judge accurately identifies a perfect form double back salto tucked with two full twists in all contests who attempt it
What are the undesired effects on performance due to influence of a variable called?
extraneous systematic variance and random error variance
True or false: a child being on their best behavior during a play observation in the clinic room but is having difficulty socially in their classroom is an example of the rosenthal effect
False- it is an example of reactive effects of testing: reacting to testing procedure differently than in a natural setting
What is it called when a test results are influenced by a large portion of the sample population withdrawing from the study?
Mortality effect
What is it called when there is an undesired effect on the DV due to some of the sample population selected being two or more standard deviations below the mean?
statistical regression effect (or being selected from the ends of the performance curve)
What is not a way to maximize the desired systematic variance (or the effect of IV on DV)?
build it into the research design
randomly select subjects that are homogenous on the DV
randomly assign subjects to groups
have a larger sample group
have a larger sample group
What is not a way to minimize random error variance?
have carefully controlled conditions
do a single subject design
short sessions to prevent fatigue or error
closely monitor the subjects
do a single subject design
How are single subject and group designs similar?
Each subject participates in all conditions of the experiments at all levels of the IV
How are single subject and group designs different?
the focus of analysis is SSd is on individual performance while for GD the analysis is on average group results in each condition
T or F: single subject designs can only be applied to ONE subject
False: Single-subject designs may be applied to only one subject or to a small number of subjects who are evaluated as separate individuals rather than as members of a larger group to be averaged together (p.144)
Why are SSD allso called time series designs?
They involve the systematic collection of a series of measurements of the dependent variable over a period of time
True or false: science is public, meaning it is accessible to other members of the scientic community for rejection or repetition
False: science is public, meaning it is accessible to other members of the scientic community for EVALUATION or repetition
True or false: examining your own knowledge in areas of research is part of the 2nd step of the scientific method (form the initial hypothesis)
False: it's part of the third step (deduction and reasoning)
What is not true of writing the research question? It should: A) be answerable by observation B) have clearly defined constructs C) devise alternative explanations D) be a repetition of another article recently published to ensure replicability E) reflect a controlled experiment
D) be a repetition of another article recently published to ensure replicability
What would make you confident that a between subject designed study with large number of subjects had groups that were equal in all pertinent attributes?
They used random sampling and random assignment
What would make you confident that a between subject designed study with smaller number of subjects had groups that were equal in all pertinent attributes?
They matched subjects into groups based on the pertinent attribues on the basis of overall distribution (aka they should attempt to ensure group equivalence)
What is this an example of?
After sustaining 'ah' for 5, 10, and 15 seconds during her oral ultrasound for her thesis, Sally's voice felt raspy and she was unable to sustain 'ah' during the 20 and 25 second trials. Her voice was shaky and she was having difficulty supporting her voice with breath.
order effect
What is this an example of?
When tested on different aspects of the SLP field such as articulation, child development, and audiological rehabilitation before the semester, Ashley scored a 40%. When she took the same test at fall break, she scored a 85%.
order effect
What is this an example of?
In her therapy session, Jules felt frustrated when the SLP started with the very hard /r/ sound and ended the session with the much easier /b/ sound. The next session, she asked her SLP to start with the /b/ sound and she felt much more confident in herself by the end of the session. She went home in a good mood.
sequence or carry-over effect
What type of research design would you suggest when studying cru de chat syndrome and why?
I would suggest a single subject design because it is a low incidence disorder. The homogeneity of the population can't be ensured and there are few individuals available for the study.
What is not a way to reduce the effects of undesired variance?
Maximize the desired systematic variance
Minimize the random error variance
Minimize the critereon error variance
Control extraneous systematic variance
Minimize the critereon error variance
Take this scenerio and tell how you would improve the hypothesis:
increasing volume of air will improve speech intelligibility in people who stutter.
tell how it will be observed/measured, define how this will be taught (where, how), include a control group, what part of speech (connected, words, etc), volume of air when? on inspiration, expiration?, how will speech intelligibility be measured?
What are some possible extraneous variables in this scenerio/hypothesis:
Children with SLI in the NYC public school system will not have better ACT scores than children without SLI in the NYC public school system.
descriptive factors: age, race, SES, reading skills. preparedness. access to prep resources. other comorbidities.
What is not an advantage of within subjects design? A) There is no problem with extraneous variables affecting a group. B) A subjects participation in an earlier condition may affect their performance in a later condition. C) When sequencing can be well controlled, the subjects act as their own control group. D) None of these above, these describe between group design.
B) A subjects participation in an earlier condition may affect their performance in a later condition.
Note
Flashcard