Reseach And Design Flashcards
Characteristics of a research study
Develop an idea into testable hypothesis
Defining the relevant variables and identifying the target population
Characteristics of a research study
Choosing an appropriate research design
Choose appropriate research strategy and specific research design
Characteristics of a research study
Selecting a sample
Determine how the sample will be selected from the population
Characteristics of a research study
Conducting the survey
Investigator conduction studies and collects and records the data
Characteristics of a research study
Analyzing to obtain data
Appropriate descriptive and inferential statistics techniques
Characteristics of a research study
Reporting the results
Prepares a report of the research results
Variable
Any characteristic behavior event or phenomenon that is capable of bearing or existing in at least two different states conditions or levels
Constant
When a characteristic is restricted to a single state or condition
Independent variable
Believe to affect or alter status of another variable
Dependent variable
Referred to as treatment or intervention is symbolize with X. Is considered the outcome of the treatment
Manipulated variables
Are considered the independent variable’s
Organismic variables
Variables that cannot be controlled by the researcher or independent variable
Nonexperimental research
Primarily to collect data about variables rather than to test hypothesis
Observational studies
Observing behavior in a systematic way often in a naturalistic settings
Case studies
Associated with an in-depth description and analysis of a single person also can entail an intensive investigation of a single institution agency community or social unit
Surveys
Administering a questionnaire either in person or by phone or through mail subject to nonresponse bias
Random assignment
Helps ensure that any observed differences between groups on the dependent variable or actually due to the effects of the independent variable
Quasi-experimental research
Does not provide the investigator with the same degree of experimental control
Ex post facto research
Involves assessing the effects of an independent variable after it has occurred
Developmental research
Conducted to assess changes that occur as function of time
Cross-sectional studies
Evaluate change over time by comparison groups of people of different ages at the same point in time
Longitudinal studies
Investigate changed by assessing people belonging to the same age group over an extended period of time
Cohort (generational) affects
Are occurring when observed differences between subjects of different ages are due to differences in experience or other factors rather then to increasing age
Cross sequential (cohort sequential) study
Combines cross-sectional and longitudinal methods by assessing members of two or more age groups at two or more different times
Simple random sample and
Every member of the population has an equal chance of being included in the sample
Stratified random sampling
When the population of interest Sperrys in terms of specific strata or characteristics that are relevant to the research hypothesis and Vesta gate or can you stratified random sample and to ensure that each stratum is represented
Cluster sampling
Sampling entails select units or clusters of individuals rather than individuals and either including all individuals in those units in research study or randomly selecting individuals from each unit
Control group
Either no treatment control group or a placebo controlled group
Placebo control groups
Are exposed to nonspecific factors that are not uniqueto a particular therapy but are common to most
Basic questions
Is there a relationship between the independent and the dependent variable?
If so is the relationship a causal one?
Three factors that can cause variability in the studies dependent variable
The independent variable or experimental variance
Systematic error due to extraneous variables and
Random error due to random fluctuations in subjects experimental conditions and methods of measurement
True experimental study
Investigators control is maximized
True experimental research
Enhances a researchers ability to maximize their ability due to the independent experimental variable
Extraneous variable or confounding variable
Source of systematic error a very bold that is a relevant to the purpose of the research
Random assignment of subjects to treatment group
Equalize is the effect of all known and unknown extraneous variables
Holding the extraneous variables constant
Selecting subjects were homogeneous
Matching subjects on extraneous variables
Making the groups equivalent in terms of the extraneous variables
Building the extraneous variables into study
Including it in the study as an additional independent variable
Internal validity
The extent that it provides accurate answers to the first two research questions
External validity
The degree that it produces an accurate answer to the third question. Can the relationship between independent independent variables be generalized to other people setting times and operations?
Basic research questions
Is there a relationship between the independent independent variable?
If so, is the relationship a casual one? Can the relationship between independent and dependent variables be generalized to other people, settings, times and operations?
A study has Internal validity
When it allows an investigator to determine if there is a causal relationship between independent and dependent variables
Threats to internal validity
Maturation
Any biological or physiological change that occurs within the subject during the course of the study as a function of time
Best controlled to include more than one group
Research
Defined as the empirical systematic investigation of the relationship between two or more variables
Threats to internal validity
History
History threatens the studies internal validity when an external event systematically affects the status of the subjects on the dependent variable
Best controlled by including more than one group
Threats to internal validity
Testing
Taking a test can alter person’s performance on the test when it is re-administered
Instrumentation
Changes in accuracy or sensitivity of the measuring device or procedures during the course of study can confound the study results
Threats to internal validity
Statistical regression
The tendency of extreme scores on a measure to regress toward the mean when the measure is readministered to the same group of people
Regression of the mean
Statistical regression is sometimes thought to be synonymous with Galton notion
Threats to internal validity
Selection
Is a threat to his studies internal validity whenever the method used to assign subjects to treatment groups result in systematic differences between the groups at the beginning of the study
Threats to internal validity
Attrition our mortality
When subjects who dropped out of one’s group differ in an important way from subjects who dropped out of other group
Threats to internal validity
Interaction with selection
Groups are initially non-equivalent, selection can act alone and/or can interact with other factors to threaten the studies internal validity
External validity
Has validity when it’s finding can be generalized to other people, settings, and conditions.
Threats to external validity
Interaction between testing and treatment
The administration of a pretest can sensitize subject to the purpose of the research study and thereby alter their reaction to the independent variable
Threats to external validity
Interactions between selections and treatment
Subjects and food in a research study can have characteristics that make them responded to the independent variable in a particular way
Threats to external validity
Reactivity – reactive arrangements
Research participants can respond to an independent variable in a particular way simply because they know their behavior is being observed
Hawthorne effect
Tendency of subjects to perform better because of attention they are receiving as a research participant or evaluation apprehension
Demand characteristics
Behavior of research participants can be altered by cues and experimental setting that inform the subject of the purpose of the study or suggest what behaviors are expected of them
Between group designs
Used the effect of different level of an independent variable are assessed by administering each level to a different group of subjects and then comparing the status or performance of the groups on the independent variable
Factorial design
Whenever a set includes two or more independent variables
Within subject design
All levels of the independent variable are administered sequential to all subjects
Single group timeseries design
One type of within subject design
Mixed designs
Combines between groups and within subject methodologies
Single subject design
Combines behavioral principles with the technique of experimental psychology to solve socially – relevant problems
Characteristics of single subject designs
Includes at least one baseline – no treatment phase and one treatment phase. Helps control any maturation affects
A B design
A. Phase and single treatment
B. Phase
Reversal design ABA, ABAB, etc.
The A.B. design can be expanded to include more than one baseline phase or more than one baseline and more than one treatment phase the extension of the A.B. design are called reversal or withdrawal designs
Multiple baseline design
Maybe unethical
Does not require withdrawal of treatment during the course of the study but instead involve sequential applying the treatment either two different behaviors of the same subject called multiple baseline across behaviors or to the same subject in a different setting
Formative evaluation
Obtaining the information needed to determine if the program is being implemented as intended or whether any modifications are needed so that the program can achieve its objectives
Summative evaluations
Entails assessing the program is affecting us in determining if the program should be continued or expanded
Continuous variable
Theoretically can take an infinite number of values on the measurement scale
Discrete variable
Can assume only a finite number of values
Qualitative variable
Places people in unordered categories
Quantitative variable
Permits comparison of people in terms of order
Nominal scale
Measurement divides variables into an ordered categories i.e. sex of sales people
Ordinal scale
More mathematically complex than a nominal scale. Divides observations into categories but also provides information on order of those categories. Likert type scale one for strongly agree seven for strongly disagree
Likert scale
A forced answer scale ranging categories from 1 to 10
Interval scales
Has a property of order as well as the property of equal intervals between successive points on the measurement scale e.g. IQ tests are usually considered a representative of the interval scale
Ratio scale
Most mathematically complex of the four measurement scales. It has properties of order and equal intervals as well as properties of an absolute zero point.makes it possible to multiply and divide racial scores to determine more precisely how much more or less of a characteristic one person has to another. Example Kelvin scale
Descriptive statistics
Used to describe or samurais a distribution of set data
Frequency distribution
Summarizing the data in terms of the number of frequency observations in each case
Cumulative frequency
Indicate the total number of observation that falls at or below each category or score
Shapes of distribution
Normal curve
Systematic bell shaped and defined by specific mathematical formula
Shapes of distribution
Kurtisis
Refers to the relative peaked nice hike or flatness of the distribution
Shapes of distribution
Platkurtic
Went to distribution is flatter
Shapes of distribution
Leptokurtic
Went to distribution is peaked
Shapes of distribution
Skewed distribution
More than half of the observations fall on one side of the distribution
Shapes of distribution
Positively skewed distribution
Most of the scores are in the negative low side of the distribution the positive tail is extended
Shapes of distribution
Negatively skewed distribution
Most scores are located on the positive high school side of the distribution and the negative tail is extended due to the presence of a few low scores
Measures of central tendency
Mode
The score category that occurs the most frequently
Measures of central tendency
Median
The score the divides a distribution in half when the data has become ordered from low to high. Useful for it it’s in sensitivity to outliers and open ended distributions where there are no specific upper and lower limits
Multimodal
Two or more scores are categories that occur equally often
Bimodal
When two scores are equal
Susceptibility to sampling fluctuations
This means that, if a large number of samples are randomly selected from the population, the mold can be expected to vary considerably from sample to sample.
Measures of central tendency
Arithmetic mean
M-X
Mean equals the sum of the means divided by number of means
Used for its lease to susceptibility to sample and fluctuation usually provides an unbiased estimate of the population mean
Measure of variability
Range
Calculated by simply subtracting the lowest score in the distribution from The highest score
Measure of variability
Variance mean square
Is a more thorough measurement of variability then the range because it’s calculations includes all the scores in the distribution rather than just the highest and lowest dance is calculated using
Measure of variability
Standard deviation
The standard deviation is calculated by taking the square root of the variance
Population parameters and sample statistics
Estimates population values based on obtain samples
Characteristics of sampling distribution
Due to the effects of random chance factors, it is unlikely that any sample will perfectly represent the population from which it was drawn
Sampling error
inaccuracies
Inferential statistics test
Indicates where the obtain samples to distichs falls in the appropriate sampling distribution
Rejection region
Region of unlikely values lights on both tales of the sampling distribution
Retention region
Regions of likely values lies in the central portion of the sampling distribution
Null hypothesis is rejected
Obtain samples statistics is in the rejection region
No hypothesis is retained
Statistical tests indicate that the sample statistic lives in the retention region
Alfa or level of significance
0.01 or 0.05 represents the sampling distribution in the rejection region and the remaining percentage represents the retention region
Type I error = alpha
Occurs when an investigator rejects a true null hypothesis
Type II error = beta
Occurs when an investigator retains a false null hypothesis
Beta
Directly calculated for a particular study depending on the Alpha
Inverse relationship
As the probability of making a Type I error increases the probability of making a Type II error Increases and vice versa
Statistical power
A statistical test enables an experimenter to reject a faults Null hypothesis
Using a one tailed test when appropriate
One tail test is more powerful than a two-tailed test
Using a parametric test
Parametric statistical tests such as a t-test or ANOVA or more powerful than nonparametric test
Homoscedacity
The variance of the population that the different groups represented are equal
Critical values and degrees of freedom
Test that allows investigator to determine whether the obtains sample value is in the rejection or retention region of the sampling distribution this is done by comparing the tested to sticks to a critical value which is the number that corresponds to the boundary that divides the sampling distribution into rejection retention
Chi– square test
Used to analyze the frequency of observation of a nominal variable test
ANOVA
Analysis of variance is used to compare two or more means
One-Way ANOVA
Is used want to study includes one independent variable and two or more dependent groups
Factor analysis of variance
An extension of the one way ANOVA that is employed want to study includes two or more independent variables
Assumptions
Use of the Pearson r and most other qualifications require that three assumptions be met
Linearity
Assumption that there is a linear relationship between the variables relationship between the X-Men why can be summarized in a straight line
Unrestricted range
Use of the Persons r is also based on the assumptions that there is an unrestricted range of scores on both variables
Homoscrdasticity
The third assumption is that the range of why scores is about the same for the values of X
Threats to internal validity
History
History threatens the studies internal validity when an external event systematically affects the status of the subjects on the dependent variable
Best controlled by including more than one group
Threats to internal validity
Testing
Taking a test can alter person’s performance on the test when it is re-administered
Instrumentation
Changes in accuracy or sensitivity of the measuring device or procedures during the course of study can confound the study results
Threats to internal validity
Statistical regression
The tendency of extreme scores on a measure to regress toward the mean when the measure is readministered to the same group of people
Regression of the mean
Statistical regression is sometimes thought to be synonymous with Galton notion
Threats to internal validity
Selection
Is a threat to his studies internal validity whenever the method used to assign subjects to treatment groups result in systematic differences between the groups at the beginning of the study
Threats to internal validity
Attrition our mortality
When subjects who dropped out of one’s group differ in an important way from subjects who dropped out of other group
Threats to internal validity
Interaction with selection
Groups are initially non-equivalent, selection can act alone and/or can interact with other factors to threaten the studies internal validity
External validity
Has validity when it’s finding can be generalized to other people, settings, and conditions.
Threats to external validity
Interaction between testing and treatment
The administration of a pretest can sensitize subject to the purpose of the research study and thereby alter their reaction to the independent variable
Threats to external validity
Interactions between selections and treatment
Subjects and food in a research study can have characteristics that make them responded to the independent variable in a particular way
Threats to external validity
Reactivity – reactive arrangements
Research participants can respond to an independent variable in a particular way simply because they know their behavior is being observed
Hawthorne effect
Tendency of subjects to perform better because of attention they are receiving as a research participant or evaluation apprehension
Demand characteristics
Behavior of research participants can be altered by cues and experimental setting that inform the subject of the purpose of the study or suggest what behaviors are expected of them
Between group designs
Used the effect of different level of an independent variable are assessed by administering each level to a different group of subjects and then comparing the status or performance of the groups on the independent variable
Factorial design
Whenever a set includes two or more independent variables
Within subject design
All levels of the independent variable are administered sequential to all subjects
Single group timeseries design
One type of within subject design
Mixed designs
Combines between groups and within subject methodologies
Single subject design
Combines behavioral principles with the technique of experimental psychology to solve socially – relevant problems
Characteristics of single subject designs
Includes at least one baseline – no treatment phase and one treatment phase. Helps control any maturation affects
A B design
A. Phase and single treatment
B. Phase
Reversal design ABA, ABAB, etc.
The A.B. design can be expanded to include more than one baseline phase or more than one baseline and more than one treatment phase the extension of the A.B. design are called reversal or withdrawal designs
Multiple baseline design
Maybe unethical
Does not require withdrawal of treatment during the course of the study but instead involve sequential applying the treatment either two different behaviors of the same subject called multiple baseline across behaviors or to the same subject in a different setting
Formative evaluation
Obtaining the information needed to determine if the program is being implemented as intended or whether any modifications are needed so that the program can achieve its objectives
Summative evaluations
Entails assessing the program is affecting us in determining if the program should be continued or expanded
Continuous variable
Theoretically can take an infinite number of values on the measurement scale
Discrete variable
Can assume only a finite number of values
Qualitative variable
Places people in unordered categories
Quantitative variable
Permits comparison of people in terms of order
Nominal scale
Measurement divides variables into an ordered categories i.e. sex of sales people
Ordinal scale
More mathematically complex than a nominal scale. Divides observations into categories but also provides information on order of those categories. Likert type scale one for strongly agree seven for strongly disagree
Likert scale
A forced answer scale ranging categories from 1 to 10
Interval scales
Has a property of order as well as the property of equal intervals between successive points on the measurement scale e.g. IQ tests are usually considered a representative of the interval scale
Ratio scale
Most mathematically complex of the four measurement scales. It has properties of order and equal intervals as well as properties of an absolute zero point.makes it possible to multiply and divide racial scores to determine more precisely how much more or less of a characteristic one person has to another. Example Kelvin scale
Descriptive statistics
Used to describe or samurais a distribution of set data
Frequency distribution
Summarizing the data in terms of the number of frequency observations in each case
Cumulative frequency
Indicate the total number of observation that falls at or below each category or score
Shapes of distribution
Normal curve
Systematic bell shaped and defined by specific mathematical formula
Shapes of distribution
Kurtisis
Refers to the relative peaked nice hike or flatness of the distribution
Shapes of distribution
Platkurtic
Went to distribution is flatter
Shapes of distribution
Leptokurtic
Went to distribution is peaked
Shapes of distribution
Skewed distribution
More than half of the observations fall on one side of the distribution
Shapes of distribution
Positively skewed distribution
Most of the scores are in the negative low side of the distribution the positive tail is extended
Shapes of distribution
Negatively skewed distribution
Most scores are located on the positive high school side of the distribution and the negative tail is extended due to the presence of a few low scores
Measures of central tendency
Mode
The score category that occurs the most frequently
Measures of central tendency
Median
The score the divides a distribution in half when the data has become ordered from low to high. Useful for it it’s in sensitivity to outliers and open ended distributions where there are no specific upper and lower limits
Multimodal
Two or more scores are categories that occur equally often
Bimodal
When two scores are equal
Susceptibility to sampling fluctuations
This means that, if a large number of samples are randomly selected from the population, the mold can be expected to vary considerably from sample to sample.
Measures of central tendency
Arithmetic mean
M-X
Mean equals the sum of the means divided by number of means
Used for its lease to susceptibility to sample and fluctuation usually provides an unbiased estimate of the population mean
Measure of variability
Range
Calculated by simply subtracting the lowest score in the distribution from The highest score
Measure of variability
Variance mean square
Is a more thorough measurement of variability then the range because it’s calculations includes all the scores in the distribution rather than just the highest and lowest dance is calculated using
Measure of variability
Standard deviation
The standard deviation is calculated by taking the square root of the variance
Population parameters and sample statistics
Estimates population values based on obtain samples
Characteristics of sampling distribution
Due to the effects of random chance factors, it is unlikely that any sample will perfectly represent the population from which it was drawn
Sampling error
inaccuracies
Inferential statistics test
Indicates where the obtain samples to distichs falls in the appropriate sampling distribution
Rejection region
Region of unlikely values lights on both tales of the sampling distribution
Retention region
Regions of likely values lies in the central portion of the sampling distribution
Null hypothesis is rejected
Obtain samples statistics is in the rejection region
No hypothesis is retained
Statistical tests indicate that the sample statistic lives in the retention region
Alfa or level of significance
0.01 or 0.05 represents the sampling distribution in the rejection region and the remaining percentage represents the retention region
Type I error = alpha
Occurs when an investigator rejects a true null hypothesis
Type II error = beta
Occurs when an investigator retains a false null hypothesis
Beta
Directly calculated for a particular study depending on the Alpha
Inverse relationship
As the probability of making a Type I error increases the probability of making a Type II error Increases and vice versa
Statistical power
A statistical test enables an experimenter to reject a faults Null hypothesis
Using a one tailed test when appropriate
One tail test is more powerful than a two-tailed test
Using a parametric test
Parametric statistical tests such as a t-test or ANOVA or more powerful than nonparametric test
Homoscedacity
The variance of the population that the different groups represented are equal
Critical values and degrees of freedom
Test that allows investigator to determine whether the obtains sample value is in the rejection or retention region of the sampling distribution this is done by comparing the tested to sticks to a critical value which is the number that corresponds to the boundary that divides the sampling distribution into rejection retention
Chi– square test
Used to analyze the frequency of observation of a nominal variable test
ANOVA
Analysis of variance is used to compare two or more means
One-Way ANOVA
Is used want to study includes one independent variable and two or more dependent groups
Factor analysis of variance
An extension of the one way ANOVA that is employed want to study includes two or more independent variables
Assumptions
Use of the Pearson r and most other qualifications require that three assumptions be met
Linearity
Assumption that there is a linear relationship between the variables relationship between the X-Men why can be summarized in a straight line
Unrestricted range
Use of the Persons r is also based on the assumptions that there is an unrestricted range of scores on both variables
Homoscrdasticity
The third assumption is that the range of why scores is about the same for the values of X
