Statistics
Response Rate
Percentage of people who actually complete the survey Ex: Mail out 1000 surveys and only 500 are returned, the rate is only 50%. -Indicates how much bias there might be in a final sample of respondents
Negative Wording
Phrasing questions with pessimism Ex: "Do you feel that the city should not approve the proposed women's shelter?" -Question wording factor
Ogive
Plot the upper boundary of each interval and the cumulative percentage
positively skewed distribution
a distribution of scores in which outliers are substantially larger (toward the right tail in a graph) than most other scores
Frequency Distribution
a summary display for a distribution of data organized or summarized in terms of how often a category, score, or range of scores occurs.
population
the set of all individuals, items,or data of interest. This is a group about which scientists will generalize. ex: if interested in students in the United States, then all U.S. students would constitute the population.
What are characteristics of the scientific method?
Data play a central role Scientists are not alone. Science is adversarial. Scientific evidence is peer reviewed
Sensitivity
Degree to which our measure can detect variations or differences in individuals true scores
Locating Proportions between two values (on opposite side of mean)
Find z scores for each value, then look at Column B for area between mean and z, and add them together to find the proportion
Locating Proportion between two values (on the same side of mean)
Find z scores for each values, then look at Column B for the area between mean and z, and subtract them, to find the proportion.
Curvilinear Relationship
Increased in the values of one variable are accompanied by systematic increases and decreases in the values of another variable. -Direction of relationship changed at least once -Also called a nonmonotonic function
Negative Linear Relationship
Increases in the values of one variable are accompanied by decreases in the second variable
Positive Linear Relationship
Increases in the values of one variable are accompanied by increases in the values of the second variable
Sample Size
Larger sample of people, the more likely that confidence interval can be reduced
Close-ended questions
Limited number of responses are given
Latin Square
Limited set of orders constructed to ensure that each condition appears at each ordinal position and that each condition precedes and follows each condition one time -balancing the conditions -this is to decrease the disadvantage of repeated measures design
Applied Research
Seeks to address issues in which there are practical problems and potential solutions Ex: Do video games such as grand theft auto increase aggression among children and young adults
Content Analysis
Systematic analysis of existing documents -Requires researchers to to devise coding systems that raters can use to quantify the information in the documents
Quantitative Variable
Varies by amount and is measured in numeric units -continuous and discrete variables Ex: Food intake in calories (continuous), pieces of food consumed (discrete)
Qualitative Variable
Varies by class and often labels for the behaviors we observe - Only discrete variables Ex: SES, categories of depression
Counterbalancing
When all possible orders of the presentation are included in the experiment -this is to decrease the disadvantage of repeated measures design
Reactivity
When awareness of being measured affects the behavior being observed or the participant reports about it -Impacts validity
Random Sample
When conducting phone interview, computer will randomly generate a list of telephone numbers (but can be biased when not all people have phones) -Probability Sampling
Order Effect
When participants know what is coming next, and this affects the dependent variable -repeated measure design disadvantage
Practice Effect
When performance on a second task is improved due to learning gained on the first task -repeated measure design disadvantage
Random Variability
When we cannot detect a relationship between variables -Can reduce random variability when by identifying systematic relationships between variables
No Relationship
Where there is no relationship between data the graph is simply a straight line
Loaded question
Written to lead people to respond in one way Ex: "Do you favor reducing public school budget?" -Question wording factor
Equation of a Regression Line
Y=bX+a
Confidence Interval
You can have 95% confidence that the true population value lies within this interval around the obtained sample result Ex: If there is a + or - 3 percentage points, and survey results indicate that 61% of student prefer to study at home, you would know that the actual population value is between 58 and 64 percent.
Frequency polygons
a dot-and-line graph used to summarize frequency of continuous data at the midpoint of each interval. -midpoint of an interval is distributed along the x-axis and is calculated by adding the upper and lower boundary of an interval and then dividing by 2.
Multimodal distributions
a distribution of scores where more than two scores occur most often or most frequently. A multimodal distribution has MORE THAN TWO modes.
Random assignment
a random procedure used to ensure that participants in a study have an equal chance of being assigned to a particular group or condition
sample
a set of individuals, items, or data selected from a population of interest -more practical and most scientific research is based upon findings in samples, not populations
Research/Scientific method
a set of systematic techniques used to acquire, modify, and integrate knowledge concerning observable and measurable phenomena
Normal Distributions (symmetrical Gaussian or bell-shaped distribution)
symmetrical distribution in which scores are similarly distributed above or below the mean, median, and mode at the center of distribution. -mean is used to summarize data
Theory
systematic body of ideas about a particular topic or phenomenon -if study confirms hypothesis, theory is supported
Ratio
Does dividing (or taking the ratio of) two numbers to represent some meaningful value?
Range
Largest Value - Smallest Value
Respect for person autonomy
informed consent
Coding
words with numeric values -useful when entering names of groups for a research study into statistical programs such as SPSS because it can be easier to enter and analyze data when group names are entered as numbers not words.
interval
- a discrete range of values within which the frequency of a subset of scores in contained
Proportion
- a part or portion of all measured data. The sum of all proportions for a distribution of scores is 1.0.
Grouped data
- a set of scores distributed into intervals, where the frequency of each score can fall into any given interval
Relative percent distribution
- a summary display that distributes the percentage of scores occurring in each interval relative to all scores distributed. -to compute relative percent: multiply the relative frequency by 100, which moves the decimal point two places to the right. -can NEVER be negative -gives the same information as a relative frequency. It is just converted to a percentage.
Cumulative Relative Frequency distribution
- a summary display that distributes the sum f relative frequencies across a series of intervals. -add each relative frequency beginning at the top or bottom of the table.
open interval
- an interval with no defined upper or lower boundary -example: we could list the interval as "126 or above" because only two values are counted.
Nominal Scales
- identify something or someone -provide no additional information -examples: Zip codes, license plate numbers, credit card numbers, country codes, telephone numbers, and Social Security numbers. -in science nominal scales are typically categorical scales that have been coded such as: race, sex, nationality, sexual orientation, hair/eye color, season of birth, marital status, or other demographic or personal information. - nominal scales answers no to all of the questions for order, difference, and ratio.
real range
- one more than the difference between the largest and the smallest number in a list of data -example: smallest value is 2 and the largest value is 170, therefore 170-2=168. The real range is 168+1+169.
Cumulative frequency distribution
- summary display that distributes the sum of frequencies across a series of intervals. - can sum from bottom up (meaningful to discuss data in terms of "less than" or "at or below" a certain value "at most") or top down.
lower boundary
- the smallest value in each interval of a frequency distribution
Mode
- the value in a data set that occurs most often or most frequently -advantage: it is simply a count; no calculations or formulas are necessary to compute a mode. -to find the mode, list a set of scores in numeric order and count the score that occurs most often
Correlation Coefficient
-Sign-tells us direction -Magnitude-the strength of the relationship -can be statistically significant/reliable if highly correlated + or - 1.00
Confounding Variables
-When the third variable is known in an non experimental method
Standard error of estimate (se)
-Whenever a single data point fails to fall exactly on regression line, there is an error on how accurately the line will predict outcome Is an estimate of the standard deviation or distance that a set of data points falls from the regression line. The standard error estimate equals the square root of the mean square residual.
Pie Charts
-a graphical display in the shape of a circle that is used to summarize the relative percent of discrete and categorical data in sectors -converting a distribution to a pie chart is simply a matter of finding the correct angles for each slice of pie. -There are 360 degrees in a complete circle, therefore, multiply each percentage by 3.6 to find the central angles of each sector (or category)
Bar chart/graph
-a graphical display used to summarize the frequency of discrete and categorical data that are distributed in whole units or classes. -like histogram except bars are separated from one another.
Quasi- experimental method
-a research study that is conducted similarly to an experiment but lacks one or both of the following: 1. The study includes a quasi- independent variable 2. the study lacks a comparison/ control group (only one group is being observed) - the variable being studied cannot be manipulated making random assignment impossible
Relative frequency distribution
-a summary display that distributes the proportion of scores in each interval. It is computed as the frequency in each interval divided by the total number of frequencies recorded. - a proportion from 0 to 1.0 that describes the portion of data in each interval. - the calculation for a relative frequency is: diving the frequency in each interval by the total frequency count.
descriptive statistics
-applying statistics to organize and summarize information -used to quantify the behaviors researchers measure -presented graphically in tabular form (tables) or as summary statistics (single values)
statistics
-branch of mathematics used to summarize, analyze, and interpret what we observed (to make sense or meaning of our observations. -commonly applied to evaluate scientific observations
Simple Frequency Distribution
-can summarize (summary display) how often each individual score occurs (ungrouped data) or how often scores occur in defined groups or intervals (grouped data)
Choosing an appropriate measure of central tendency
-depends largely on the type of distribution and the scale of measurement of the data.
1. Changing and existing score will change the mean
-every score in a distribution affects the mean; therefore, changing any existing score in a distribution will change the value of the mean
outliers
-extreme scores that fall substantially above or below mot of the scores in a particular data set
Histograms
-graphical display used to summarize the frequency of continuous data that are distributed in numeric intervals (grouped). -distribute the intervals along the horizontal scale (x-axis) and list the frequency of scores in each interval on the vertical scale (y-axis)
2. Adding a new score or removing an existing score will change the mean, unless that value equals the mean
-if the new score (x) is less than the mean the mean will decrease -if the new score (x) is greater than the mean the mean will be increased -deleting a score below the mean will increase the value of the mean -deleting a score above the mean will decrease the mean -the only time that a change in a distribution of scores does NOT change the value of the mean is when the value added or removed is exactly equal to the mean.
describing interval and ratio scale data
-mean is used for data that can be described in terms of the DISTANCE that scores deviate from the mean -the mean is an appropriate measure of central tendency used to describe interval and ratio data because differences between two scores are meaningfully conveyed for data on an interval or ratio scale only.
three measures of central tendency:
-mean, median, mode
Correlational Method
-measure pairs of scores for each individual. -This can determine whether a relationship exists between variables, but lacks the appropriate controls needed to demonstrate cause and effect. -we can examine the extent to which two variables change in a related fashion.
continuous variable
-measured along a continuum at any place beyond the decimal point. -can be measured in fractional units -example: consider that Olympic sprinters are timed to the nearest hundredths place, but the Olympic judges wanted to clock them to the nearest millionths place, they could.
Discrete variable
-measured in whole units or categories that are not distributed along a continuum -example: the numbers of brothers and sisters you have and your family's socioeconomic class (lower class, middle class, upper class)
Ratio Scale
-measurements that have a true zero and are distributed in equal units -examples: counts and measures of length, height, weight, and time
Interval Scales
-measurements that have no true zero and are distributed in equal units -equidistant scales (scale with intervals or values distributed in equal units) and no true zero - example in behavioral science: the rating scale
Percentile Points
-measures of the relative position of individuals or scores within a larger distribution -the value of a score on a measurement scale below which a specified percentage of scores in a distribution fall.
describing ordinal scale data
-median is used to describe ranked or ordinal data that convey direction only. -because the distance (or deviation) of ordinal scale scores from their mean is not meaningful, the median is appropriate to use.
quasi-independent variable
-preexisting variable that is often a characteristic inherent to an individual, which differentiates the groups or conditions being compared in a research study. Because the levels of the variable are preexisting, it is not possible to randomly assign participants to groups. ex: sex (male or female) because sex cannot be randomly assigned
Scales of measurement
-rules for how the properties of numbers can change with different uses -nominal, ordinal, interval, and ratio -characterized by three properties: order, difference, and ratio.
central tendency
-statistical measures for locating a single score that is most representative or descriptive of all scores in a distribution (single values that have a "tendency" to be near the "center" of a distribution. -stated differently for populations and samples
cumulative percents distribution
-summary display that distributes the sum of relative percents across a series of intervals -sum the relative percent in each interval, following the same procedures for adding as we did for the cumulative relative frequencies.
3. adding, subtracting, multiplying, or dividing each score in a distribution by a constant will cause the mean to change by the constant
-when each value is changed by the same constant, the mean is also changed by the constant. (applies when every score in the distribution is changed by the same constant) - multiplying each value in the original distribution by a certain value changes the mean by a multiple of that certain value. (applies when every score in the distribution is changed by the same constant.
Haphazard Sampling
"Convenience Sampling" Ex: Select any sample of students from your school in any way that is convenient -More likely to introduce biases and therefore might not be an accurate representation of the population of all students -Nonprobability Sampling
Describing skewed distributions
-occurs whenever a data set includes a score or group of scores that fall substantially above (positively skewed) or substantially below (negatively skewed) most other scores in a distribution. -in a normal distribution the mean and mode are equal, in a positively skewed distribution the mean is greater than the mode; in a negatively skewed distribution the mean is less than the mode. -location of the median is unpredictable and can fall on either side of the mode depending on how the scores are distributed. -outliers distort the value of the mean, making it a less meaningful measure for describing all data in a distribution.
Experimental Control
All extraneous variables are kept constant -Variable that is held constant cannot be a confounding variable
SIQR (Semi-interquartile range)
(Q3-Q1)/2 Is a measure of half the distance between the upper (Q3) and the lower quartile (Q1) of the data set, and is computed by dividing the IQR in half
5. Th sum of the squared differences of scores from their mean is minimal.
-suppose that we want to measure how far scores deviate from the mean. We could subtract each score from the mean and sum the deviations, but that will always produce a result equal to 0. If we did this, then we would erroneously conclude that scores do not deviate from their mean. The solution to obtain a value greater than 0 is to square each deviation before summing.This produces the smallest possible positive number greater than 0, where larger outcomes indicate that scores deviate further from their mean. -The notation for describing the sum of squared differences of scores (x) from their mean (M) is: sum of (x-M) squared. -subtracting scores from any constant other than the mean will produce a greater value. -if you substitute any positive or negative value other than the mean, you will always obtain a greater solution.
Weighted Mean
-the combined mean of two or more groups of scores in which the number of scores in each group is disproportionate or unequal -formula for the weighted mean: weighted sum divided by combined n -to compute the weighted mean, we find the product, M x n, for each sample. This gives us the weight for the mean of each sample. By adding these products, we arrive at the weighted sum. Then, divide the weighted sum by the combined sample size (n), which is computed by adding the sample sizes in the denominator
upper boundary
-the largest value in each interval of a frequency distribution
Using the median to describe data
-the median is typically used to describe data distributions that are skewed and measures on an ordinal scale
Median
-the middle value in a distribution of data listed in numeric order -to find the median list a set of scores in numeric order and compute this formula: n+1 divided by 2. -can be estimated by a cumulative percent distribution. Because the median is located in the middle of a distribution, it is approximately at the 50th percentile of a cumulative percent distribution.
Describing nominal scale data
-the mode is used to describe nominal data that identify something or someone, nothing more. -not a quantity so it doesn't make sense to use the mean and median to describe data. The mode is used instead. -anytime you see phrases such as: most often, typical, or common, the mode is being used to describe these data.
sample size
-the number of individuals who constitute a subset of those selected from a larger population. -represented by a lowercase "n"
population size
-the number of individuals who constitute an entire group or population -represented by a capital "N"
Percentile Rank
-the percentage of scores with values that fall below a specified score in a distribution.
Interval width (class)
-the range of values contained in each interval of a grouped frequency distribution - to find this we divide the real range by the number of intervals chosen. -recommended numbers of intervals is 8-12 or 5-20. (anything less provides too little summary; anything more is often too confusing)
Operational Definition
-the set of procedures used to measure or manipulate it -A variable must be operationally defined in order to be studied empirically -Ex:"bowling skills" person's average score over the past 30 games
Mean (arithmetic mean or average)
-the sum of a set of scores in a distribution, divided by the total number of scores summed. -most commonly reported measure of central tendency -population mean formula: sum of N scores(x) divided by N (identified by the greek letter mu) -sample mean formula: sum of n scores(x) divided by n (identified by an italicized M or X bar) -often referred to as the "balance point" in a distribution
interval boundaries
-the upper and lower limits for each interval in a grouped frequency distribution
4. The sum of differences of scores from their mean is zero
-think of the mean as the balance point of a distribution of scores. What logically follows from this is to think of the mean as a zero point for a distribution as well. -It is the only constant you can subtract from every score in a distribution, where the sum of the differences is equal to zero. -The notation for describing the sum of the differences of scores from their mean is sum of (x-M) -Only when the mean is subtracted from each score in a distribution is the sum of the differences equal to zero
inferential statistics
-used to infer that observations made with a sample are also likely to be observed in the larger population from which the sample was selected.
using the mean to describe data
-usually used to describe normally distributed data and measures on an interval or ratio scale.
Quantitative variable
-varies by amount. This variable is measured numerically and is often collected by measuring or counting. -continuous and discrete variables can be quantitative -example: we can measure food intake in calories (a continuous variable), or we can count the number of pieces of food consumed (discrete variable)
qualitative variable
-varies by class. -This variable is often represented as a label and describes nonnumeric aspects of phenomena. -often labels for the behaviors we observe, therefore, only discrete variables can fall into this category. -example: socioeconomic class (working class, middle class, upper class) is discrete and qualitative.
Characteristics of Mean
1) Changing existing score will change mean 2) Adding or remaining existing score will change mean unless that value equals the mean -Deleting a value below the mean will increase value of mean -Deleting a value above the mean will decrease the value of the mean 3) Adding, Subtracting, Multiplying or Dividing each score in a distribution will cause the mean to change in that constant 4) Sum of difference of scores is 0 5) Sum of squared differences of scores from their mean is minimal
3 Assumptions for Linear Correlations
1) Normality 2) Linearity 3) Homoscedacsity
4 Restrictions on appropriateness/validity of Pearson r linear correlations
1) Outliers 2) Curvilinearity 3) Restricted (truncated) range 4) Reliability of measures
Characteristics of Standard Deviation
1) Standard Deviation is always positive 2) SD is used for quantitative data 3) SD is most informative with the mean 4) The value of SD is affected by value of each score in a distribution
three rules for constructing a histogram
1. A vertical rectangle represents each interval, and the height of the rectangle equals the frequency recorded for each interval. 2. The base of each rectangle begins and ends at the upper and lower boundaries of each interval. 3. Each rectangle touches adjacent rectangles at the boundaries of each interval
Empirical Rule
1. At least 68% of all scores lie within one SD of the mean 2. At least 95% of all scores lie within two SD of the mean 3. At least 99.7% of all scores lie within three SD of the mean
Five Key Characteristics of the Mean
1. Changing and existing score will change the mean 2. Adding a new score or removing an existing score will change the mean, unless that value equals the mean 3. adding, subtracting, multiplying, or dividing each score in a distribution by a constant will cause the mean to change by the constant 4. The sum of differences of scores from their mean is zero 5. Th sum of the squared differences of scores from their mean is minimal.
Two characteristics of the bar chart/graph
1. Each class or category is represented by a rectangle 2. Each rectangle is separated along the x-axis
four rules for creating a simple frequency distribution
1. each interval is defined (it has a lower and upper boundary). Intervals such as "or more" or "less than" should not be expressed. 2. each intervals equidistant (the interval width is the same for each interval). 3. No interval overlaps (the same score cannot occur in more than one interval) 4. all values are rounded to the same degree of accuracy measured in the original data
Nuremberg Code
10 rules of research conduct that would help prevent future atrocities
Tuskegee Syphilis
399 African American not treated for syphilis in order to track long term effects of the disease-unethical 1932-1972
Slope of a regression line (b)
A _______ of a straight line is used to measure the change in Y relative to the change in X. When X and Y change in the same direction, the ____ is positive. When X and Y change directions, the ______ is negative.
Quota Sampling
A researcher uses this technique and chooses a sample that reflects numerical composition of various subgroups but are not randomly sampled Ex: Ensure that sample of students are 19% freshmen, 23% sophomore, 26% juniors, 22% seniors and 10% grad students and these are the percentages of the classes in total population -Collect data by using the haphazard techniques -Nonprobability Sampling
Purposive Sampling
A sample of people that meet some predetermined criteron Ex: At a large movie complex, researchers are asking customers to fill out questionnaires about one or more movies under the age of 30 or under -Nonprobability Sampling
Outlier
A score that falls substantially above or below most other score IQR is best to use when there is an outlier in regards to variability and the median is best to use in regards to central tendency
Interview
A social interaction with participants -can be face-to-face i -over the telephone -focus group
Independent Variable
A variable that is manipulated
Panel Studies
A way to study changes over time is by surveying the same people two or more points in time -Important to use when research questions addresses the relationship between one variable "time 1" and another variable at some time later "time 2".
Structure of empirical journal
Abstract, intro, methods, results, discussion, references
Normality
All data points are normally distributed -for linear correlation between two factors, the assumption of ______ requires that a population of X and Y scores for two factors forms bivariate (two-variable) normal distribution -Assumption for linear correlation
Participant Observation
Allows researcher to observe the setting from the inside, he or she may be able to experience events in the same way as natural participants -issue with this observation is that observations will likely be biased and conclusion will lack objectivity
Ordinal Scales
Allows us to rank order the levels of the variable being studied -Categories can be ordered from first to last Ex: Letter grades
Case Study
An observational method that provides a description of an individual, business, school or neighborhood
IACUC Research with non-human animals
Animal research is very important but need to follow the rules of ethical guidelines; proper housing, feeding, cleanliness and health care
Coercion
Anything that you do that make the participant feel like they have to participant Ex: monetary reinforcement
Test-retest
Assessed by measuring the same individuals at points in time Ex: IQ test can be assessed on one day and then again a week later
Internal Consistency
Assessment of reliability using responses at one point -Since all items measure the same variable, they should yield similar or consistent results
Integrity
Being truthful and accurate honest -APA code
Regression Line
Best-fitting straight line to a set of data points. It is a lines that minimizes the distance of all data points that fall from it
Declaration of Helsinki
Broader application of the Nuremberg Code that was produced by the medical community
mean
Called the arthimetic mean or average -sum set of scores in a distribution, divided by the total number of scores summed
Rating Scales
Can be a 5 or 7 point scale ranging from strongly agree to strongly disagree -More quantifiable and preferable for research
Participant Variables
Characteristics of individuals such as age, gender, ethnic group, nationality, birth order, personality, or marital status -nonexperimental by definition
weighted mean
Combined mean of two or more groups of scores in which the number of scores in each group is disproportionate or unequal
Population
Composed of all individuals of interest to the researcher
Risk benefit analysis
See if there is more risks than benefits or more benefits than risks -physical harm -stress -confidentiality
Reliability
Consistency or stability of a measure of behavior Ex: Professor is reliable because he always comes to class on time -This measure does not fluctuate from one reading to the next
"Yea-saying (Acquiescence) Nay-saying"
Consistently agreeing or disagreeing with a set of related questions phrased in both standard and reversed formats is an indicator to this type of response set
Belmont Report
Current ethical guidelines for both behavioral and medical researcher -3 principles
Fatigue Effect
Deterioration in performance from first to second task due to boredom, tiredness or distraction -repeated measure design disadvantage
Order
Does a larger number indicate a greater value than a smaller number?
Difference
Does subtracting two numbers represent some meaningful value?
Attrition/Mortality
Dropout factor in experiment
Probability Sampling
Each member of the population has a specifiable probability of being chosen
Independent groups design
Each participant are assigned to each condition by random assignment -Will prevent any systematic biases and can be considered equal due to random assignment
Randomization
Ensures that an extraneous variable is just as likely to affect one experimental group as it is to affect the other group -Researcher assigns participants to the two groups in a random fashion (random assignment)
Fidelity and Responsibility
Establishing a relationships of trust with those whom they work -being professional -APA codes
Simple Random Sampling
Every member of the population has an equal probability of being selected for the sample -Probability Sampling -If there is 1000 students, each student has a chance of out the 1000 to be selected
Concurrent Validity
Examines the relationship between measure and a criterion behavior at the same time -Ability to distinguish between groups that it should theoretically be able to distinguish between Ex: measure of shyness on salespeople who make cold calls to customers (score lower on shyness) versus salespeople whose customers must make contact with the actual company
Outliers
Extreme scores that fall substantially above or below most scores in a particular data set -restriction on appropriateness of correlation
Fraud
Fabrication of data -Detected when other scientist cannot replicated -Sir surreal bert
Locating Scores (Example, top 10%, or lower 20%):
Find the p value that equals the percent-age (Example, top 10% = .1000 and lower 20% = .2000). Then locate the Z from column C. Fill in the numbers to the z score equation and solve. -top, right =positive z score (top 10%) -low,bottom, left=negative z score (bottom 20%) -this will affect your final answer
Survey
Used to assess people's thoughts, feelings and opinions -surveys can be mailed -surveys can be online
Matched pair design
First match people on a participant variable such as age or personality trait, and the matching will be either the dependent variable or a variable that is strongly related to the dependent variable Ex: In a learning experiment, participants are matched on basis of scores on a cognitive ability measure or gpa Disadv- can be time consuming and costly
Qualitative Research
Focuses on people behaving in natural settings and describing their world in their own words -Emphasizes on collecting in-depth information on a relatively few individuals or within a limited setting
Ratio Scales
Have an absolute 0 that indicates the absence of the variable being measured and are equal in size -Used in behavioral sciences that involve physical measures being studied like time measures Ex: reaction time, rate of responding, and duration of response
Nominal Scales
Have no numeric or quantitative properties -Used to identify -catergorical variables -IV is often nominal Ex: gender, zip code, license plate
Time-intervaling
Having a time in between conditions to avoid the fatigue effect -this is to decrease the disadvantage of repeated measures
Research Questions
Identifying and describe broad topic that is being investigates
Measurement Error
If it is unreliable and does not have a accurate indication
Directionality
In non-experimental method it is hard to determine which variable causes the other
Third Variable
In non-experimental method there is a danger that no direct causal relationship exists between the two variables -Also known as a spurious relationship -An extraneous variable that may be responsible for observed relationship between both variables
Split-Half Reliability
Indicator of internal consistency reliability -Correlation of the total score on one half of the test with the total score on the other half
Cronbach's Alpha
Indicator of internal consistency reliability -Provides us with the average of all possible split-half reliability coefficients -Scores on each item are correlated with scores on every other item
Interval Scales
Intervals between numbers are equal in size but have no true zero -Generally have 5 or more quantitative levels Ex: surveys, questionnaires
Experimental Method
Involves direct manipulation and control of variables -Researcher directly manipulates the variable of interest and then observes the response
Archival Research/Data
Involves using previously compiled information to answer research questions
Variability
Is a measure of dispersion or spread of scores in a distribution and ranges from 0 to infinity -It includes all scores versus two extreme scores like the range
Pearson correlation coefficient/correlation coefficient
Is a measure of the direction and strength of linear relationship of two factors in which the data for both factors are measured on an interval or ratio scale of measurement
Standard Deviation/Root Mean Square Deviation
Is a measure of variability for the average distance that scores deviate from their mean -It is calculated by taking the square root of the variance
Sampling Frame
Is the actual population of individuals (or clusters) from which a random same will be drawn -Does not represent population -can introduce bias
Coefficient of Determination (r2 or R2)
Is the mathematically equivalent to eta-squared and is used to measure the proportion of variance of the one factor (Y) that can be explained by known values of a second factor (X) -r2=(.744)2=.553
Sums of Square
Is the sum of squared deviations of scores from their mean. The SS is the numerator in the variance formula -SS/N
Sample Space
Is the total number of possible outcomes that can occur in a given random event (denominator in fraction)
Criterion Variable or to-be predicted variable (Y)
Is the variable with unknown values that can be predicted or estimated, given the known values of the predictor variable
Coding System
Is used to measure behavior in a natural setting Ex: Bakeman & Brownlee social behavior of young children. Had to identify "free play" as unoccupied, solitary play, together, parallel play, and group play
Why can't correlation imply causality?
It can only imply the direction and strength of a relationship, but no cause and effect -reverse casuality (directionality): casuality between two factors can go in either direction-can be due to third or confounding variables
Convergent Validity
Scores on the measure are related to other measures of the same construct Ex: New depression scale is related to the already established Beck Depression
Locating Proportions above the mean
Look at Column C (the tail)
Locating Proportions below the mean
Look at Column C (the tail)
Variance
Measure of variability for average squared distanced that scores deviate from the mean
Continuous Variable
Measured along a continuum at any place and beyond the decimal point Ex: Olympic Sprinter's time
Discrete Variable
Measured in whole units and not along a continuum -nominal scales -ordinal scales
Solomon four-group design
Merging the pretest-posttest design with the postest design to maximize the advantages and avoid the pitfalls of using the pretest experiments
Posttest-only design
Must have two equivalent groups of participants, has to introduce the independent variable and measure the effect of the independent variable on the dependent variable. -has a high degree of internal validity -Experimental Design
Advantage of multiple methods
No study is a perfect test of a hypothesis, but multiple studies using multiple methods, that leads to the same conclusion increase our confidence in our finding
Debriefing
Occurs after the completion of a study, tells them what the purpose of the study was
Deception
Occurs when there is a misrepresentation of information about the nature of the study -Milgram
Sample
Participants from a population of interest
Stratified Random Sampling
Population is divided into subgroups (strata) and random sampling techniques are then used to select sample members from each group. Ex: Survey of sexual attitudes, might be divided by age, gender and amount of education -surveyed individual -Probability Sampling
Carryover Effect
Possible that effects on first condition to be carried over to the second condition -repeated measure design disadvantage
Informed Consent
Potential participant should be provided with all information that might influence their active decision whether or not to take part in the study
Restricted (truncated) range
Problem that arises when the range of the data for one or both correlated factors in a sample is limited or restricted, or smaller that the range of data of the general population -This is a restriction of appropriateness of correlation
IQR (Interquartile Range)
Q3-Q1=IQR Is the range of values between the upper (Q3) and the lower (Q1) quartiles of a data set -This is used when there is an outlier in the range
Questionnaires
Questions are presented in written format and the respondents write their answers and can be anonymous -can be administered in person to groups or individual, through the mail, or the internet
Simplicity
Questions asked in a survey should be easy to understand -Question wording factor
Doubled-barreled question
Questions that ask two things at once Ex: "Should senior citizens be given more money for recreation centers and food assistance programs?" -Question wording factor
True score
Real score of the variable -said to be reliable
Construct Validity
Refers to adequacy of the operational definitions of variables -Adequacy of the operational definition -Truly reflects the theoretical meaning of the variable
Internal Validity
Refers to our ability to accurately draw conclusions about causal relationships -Must have temporal precedence:manipulating IV and then observed whether it has effect on DV -Must have covariation: exp. group shows conditioned effect vs control group and no effect -Must eliminate alternative explanations
Systematic Observation
Refers to the careful observation of one or more specific behaviors in a particular setting -Observations are quantifiable and the researcher had developed prior hypotheses about the behaviors -Can be used in natural settings and laboratory settings -developed a coding system to measure behavior prior to the study
Non-Experimental Method
Relationships are studied by making observations or measures of variables of interest -Variables are observed as they occur naturally -Correlational Method -Not ideal when we want to know cause and effect
Discriminant Validity
Scores on the measure are not related to other measures that are theoretically different Ex: Show how the Head Start program is not similar to other early childhood programs that do not label themselves as Head Start Programs
Graphic rating scales
Requires a mark along a continuous 100-millimeter line that is anchored with descriptions at each end -a ruler is placed on the line to obtain the score on a scale that ranges from 0 to 100
Alternate forms reliability
Requires administering two different forms of the same test to the same individuals at two points in time -Can be used to decrease artifically high correlation due to remembering IQ measure in test-retest
Exempt/No risk
Research in which there is no risk Ex: surveys, questionnaire -must be reviewed -IRB conditions
Naturalistic Observations
Researcher makes observations of individuals in their natural environment (the field) -goal is to provide a complete and accurate picture of what occurred in the setting
Cluster Sampling
Researchers can identify "clusters" of individuals and then sample each of these groups Ex: Classes being taught, randomly sample these classes and then have all the students from the particular class complete survey -surveyed in group -Probability Sample
Placebo Effect
Researchers often administer a neutral treatment to the control group, usually a sugar pill, to examine differences between IV and DV
Semantic Differential Scale
Respondents are asked to rate any concepts-persons, objects, behaviors, ideas-on a series of bipolar adjectives using a 7-point scale
Open-ended questions
Respondents are free to answer in any way they like
Social Desirability
Response set leads the individual to answer in the most direct acceptable way-the way that "most people" are perceived to respond or the way that would reflect most favorably on the person -"faking good" -Most common response set
Institutional Review Board
Responsible for the review of research conducted within the institution
Repeated Measures Design
Same individual participates in both conditions -Adv: fewer participants, extremely sensitive to finding statistical significance in differences between groups -Major Disadv: Presentation in a particular sequence
Annual Survey
Same questions each year able to track changes over time
Steps to finding percentile rank
Step 1: Identify the interval within which a specified percentile point falls Step 2: Identify the real range for the interval identified. Step 3: Find the position of the percentile point within the interval -To identify the position of the percentile point, first find the distance of the percentile from the top of the interval. Next, divide the number of points the percentile is from the top by the total range width of the percentages. (example: If 75% is 10 points from the top of the interval and the total range is 24 you divide 10/24 of the total interval then multiply the fraction by the width of the real range, which is 9 points on this example 10/24 X 9= 3.75 points) This means the position of the percentile point is 3.75 points from the top of the interval Step 4: Identify the percentile point -In the example: the top of the interval is 98.5. Subtract 3.75 from that value to identify the percentile point at the 75th percentile (98.5-3.75=94.75 thus, the percentile point at the 75th percentile is 94.75.
Greater than minimal risk
Subject to thorough review -informed consent needed -debriefing -confidentiality safeguards -IRB conditions
Snowball Sampling
Surveying one person, and then have that person identify another person and so on and so forth -interested in a specialized population -Nonprobability Sampling
Pearson Product-moment correlation coefficient
Symbolized as r and can range from 0.00-1.00 or 0.00-(neg)1.00 -0.00 no relationship -1.00 positive relationship (strong) both (x&y) increase -(neg)1.00 negative relationship (strong) (x) increase and (y) decrease -should be .80 for strong reliability -referred also to reliability coefficient
Hawthorne Effect
Tendency of some people to perform better and work harder when they are participants in an experiment
Quantitative Research
Tends to focus on specific behaviors that can be easily counted and investigations are generally include larger samples
Hypotheses
Tentative idea or question waiting to for evidence to support or refute it (testable prediction)
Field Experiment
The IV is manipulated in a natural setting -Good for external validity/low for internal validity -Disadv. is that researcher loses ability to directly control many aspects of the situation
Y intercept (a)
The _______ of a straight line is the value of the criterion variable (Y) when the predictor variable (X) equals 0
Homoscedascity
The assumption of constant variance among data points. There is an equal (homo) variance or scatter (scedacsity) of data points dispersed along the regression line -(Hetero) no equal variance
Linearity
The assumption that the best way to describe a pattern of data is using a straight line -restriction would be curvilinearity
Curvilinearity
The best fitting regression surface of Y on X is not a straight line. The mean falls on a curved line rather than a straight line-violates linearity -This is a restriction of appropriateness of correlation
Sampling Error/Margin of Error
The confidence interval gives you information about the likely amount of error.
Content Validity
The content of the measure is linked to the universe of content that defines the construct Ex: Depression is defined by mood and by cognitive and physiological symptoms. Each item would have links to each of the symptoms that define the depression construct as a whole -Reflect meaning of construct
Apa Ethics Codes
The ethical principals of psychologist and code of conduct
Face Validity
The evidence of validity is that the measure appears "on the face of it" to measure what it is supposed to measure -Ex: Measure of a new depression scale has items like "I feel sad" or "I feel down" or "I cry alot." -Reflect meaning of construct
Inter-rater Reliability
The extent to which 2 researcher observing the same event to see if they're consistent -indicator is Cohen's kappa
External Validity
The extent to which our results of a study can be generalized to other populations and settings -Field Studies
Predictive Validity
Uses a measure to predict some future behavior Ex:Measure of depression predicts future diagnoses of depression -This is a criterion because it is based off of behavior
Probability
The frequency of times an outcome occurs divided by the total number of possible outcomes - Identify the likelihood of a random event, not fixed - Varies between 0 and 1, can never be negative
Spearman rank-order correlation coefficient (rs) or Spearman's rho
The measure of strength and direction of the linear relationship between two ranked factors on an ordinal scale of measurement
Point-Biseral Correlation Coefficient (rpb)
The measure of the direction and strength of the linear relationship of one factor that is continous (interval or ratio scale of measurement) and a second factor that is dichotomous (on a nominal scale of measurement)
median
The middle value in a set of numerically ordered data
Degrees of Freedom for Sample Variance
The numbers of scores in a sample that are free to vary -All scores except 1 are free to vary in a sample -n-1
Pretest-Postest Design
The only difference from that of a posttest design is that a pretest is given before the experimental manipulation is introduced -this design makes sure that the groups, were in fact, equivalent at the beginning of the experiment -adv can determine possibility of dropouts (attrition), can help up focus on the change from pretest to postest -disad. can be time consuming, can sensitize part. to what you are studying
Pearson r (Product-Moment)
The pearson correlation coefficient is used to describe the relationship between 2 factors on an interval or ratio scale
Nonprobability Sampling
The probability of any particular member of the population being chosen is unknown
Nonverbal scales for children
Uses drawings of faces for ratings such as a toy
Mode
Value that occurs the most in a set of data -reported with mean and median
Response Set
The tendency to respond to all questions from a particular perspective rather than to provide answers that are directly related to the questions
Dependent Variable
The variable that is being measured
Predictor Variable or known variable (X)
The variable with values that are known and can be used to predict values of another variable
Minimal Risks
There is no harm to the participants, no greater than the risk encountered in daily life -debriefing/ethical concerns are important -generally no informed consent -IRB conditions
Quasi Independent Variable
There is no random assignment and no control group
Phi Correlation Coefficient (Cramer's Phi)
These are two dichotomous factors, that can only take on two values, are typically categorical and are therefore measured on a nominal scale -to measure strength and direction of the linear relationship between the two dichotomous factors
Labeling Response Alternative
This type of scale assumes that the middle alternative is "neutral" point half way between the endpoints.
Discriptive Studies
To describe behaviors, causal inferences
Selection Differences
To have equivalent groups of participants in the posttest design, the people selected cannot differ in any systematic way
Basic Research
Tries to answer fundamental questions about the nature of behavior/knowledge -concerning theoretical issues Ex: Is extraversion related to sensation seeking
Population Parameter
a characteristic (usually numeric) that describes a population ex: we want to test if a new learning tool can improve learning in this population; this characteristic (learning) in the population is the population parameter (learning will be the characteristic that will be measured, but not in the population)
sample statistic
a characteristic (usually numeric) that describes a sample -value that is measured in the study -measured to estimate the population parameter
Operational definition
a description of some observable event in terms of the specific process or manner by which it was observed or measured. -ex: define exam performance as a score between 0 to 100 on a test
Unimodal distribution
a distribution of scores in which one score occurs most often or most frequently. A unimodal distribution has ONE mode. - mode can be used with the mean to describe skewed distributions
negatively skewed distribution
a distribution of scores in which outliers are substantially smaller (toward the left tail in a graph) than most other scores.
bimodal distribution
a distribution of scores in which two scores occur most often or most frequently. A bimodal distribution has TWO modes. -mean and median located between the two modes.
Nonmodal distribution
a distribution of scores where all scores occur at the same frequency. A nonmodal distribution has NO mode -essentially a straight line with the frequency of each score being the same -mean and median are located toward the center of the nonmodal distribution.
Ogive
a dot-and-line graph used to summarize the cumulative percent of continuous data at the upper boundary of each interval -The y-axis of an ogive always ranges from 0% to 100% of the data -plotting at the upper boundary of each interval is necessary because this point represents or contains all the scores in that interval.
Beneficence
benefits should outweigh risks -In apa is beneficence/nonmaleficence
temporal precedence
cause precedes the effect -necessary to show casuality
Survey Archives
consist of data from survey that are stored on computers and available to researchers who to analyse them
describing modal distributions
distribution of scores in which one or more scores occur most often or most frequently
three research methods
experimental method quasi-experimental method correlational method
Justice
fairness in receiving benefit of the research
Frequency Polygon
is a dot and line graph used to summarize the frequency of continuous data at the midpoint of each interval -add the lower and upper boundary and divide it by 2
To construct a bar chart/graph...
list the whole units or categories along the x-axis, and distribute the frequencies along the y-axis
Three requirements of control (experimental method)
manipulation (of variables that operate in an experiment) randomization (of assigning participants to conditions) - to meet the requirement of randomization, researchers must use random assignment to assign participants to a group comparison/control (a control group)
Sector
particular portion of a pie chart that represents the relative percent of a particular class or category
Squared correlation coefficient r2
percent or proportion of variablity in Y (criterion outcome) that is associated with X (predictor) -If r and r squared are high in value, it does not mean they are related, there can be third variables or confounding variables
Plagarism
refers to misinterpreting another's work at your own
To meet the requirement of randomization...
researchers must use random assignment to assign participants to a group. In order to do this, a researcher must be able to manipulate the levels of independent variables (IV) to create groups
Ordinal Scales
scale that conveys only that some value is greater or less than another value. -examples: finishing order in a competition, education level, and rankings. -Ordinal scales answer yes to the question of order, but it answers no to the questions of difference or ratio.
statistics are often used in context of...
science -in the behavioral sciences, science is specifically applied using the research method
Step to construct a simple frequency distribution for grouped data
step 1: find the real range step 2: find the interval width step 3: construct the frequency distribution
Using the mode to describe data
the mode is typically used to describe data with modal distributions and measures on a nominal scale
experiment (experimental method)
the use of methods and procedures to make observations in which the researcher fully controls the conditions and experiences of participants by applying three required elements of control to isolate cause-and-effect relationships between variables
Independent Variable
the variable that is manipulated in an experiment. This variable remains unchanged (or "independent") between conditions being observed in an experiment. It is the "presumed cause". The specific conditions of an IV are referred to as the levels of the independent variable.
dependent variable
the variable that is measured in each group of a study and is believed to change in the presence of an independent variable. It is the "presumed effect" -require an operational definition
Alternative Explanation
there is no other explanation for the effect necessary to show casuality
Covariation
when the cause is present and so is the effect when it is not present there is no effect -necessary to show casuality Ex: watch violent equal aggression
true zero
when the value 0 truly indicates nothing on a scale of measurement -example: temperature
z-score
z = x - M divided by SD - Mean of 0, SD of 1