Week 2 Flashcards
Key components to a statistical investigation
Planning the study, examining the data, inferring the data, drawing conclusions, distributional thinking
Examining the data
The most fundamental principle of statistics is that data may vary. The pattern of that variation is crucial to capture and understand. Values of a variable vary. Analyzing the pattern of variation, called the distribution of the variable, often reveals insights
Statistical significance
Even when patterns in data are found, there is often still uncertainty in various aspects of the data. There may be potential for measurement errors (even body temperature can fluctuate by almost 1 degree over the course of the day) We may only have a “snapshot” of observations from a more long-term process or only a small subset of individuals from the population of interest
P-value
The probability of observing a particular outcome in a sample, or more extreme, under a conjecture about the larger population or process. Tells you how often a random process would give a result at least as extreme as what was found in the actual study, assuming there was nothing other than random chance at play
Level of significance
A result is statistically significant if it is unlikely to arise by chance alone. If the p-value is smaller than the cut-off value, then we reject the hypothesis that only random chance was at play
sample
The collection of individuals on which we collect data. Sample from a larger group of individuals (the population) in such a way that conclusions from the sample can be generalized to the larger population
Generalized
Related to whether the results from the sample can be generalized to a larger population
population
A larger collection of individuals that we would like to generalize our results to
Random sample
using a probability-based method to select a subset of individuals for the sample from the population. Involves numbering every member of the population and then using a computer to randomly select the subject to be surveyed
Margin or error
The expected amount of random variation in a statistic; often defined for 95% confidence level
Non-random samples
Often suspect to bias, meaning the sampling method systemically over-represents some segments of the population and under-represents others. Consider other sources of bias, such as individuals not responding honestly
cause and effect
Related to whether we say one variable is casing changes in the other variable, versus other variables that may be related to these two variables
Randomly assigning
using a probability-based method to divide a sample into treatment groups. Apply the probability model to approximate as a p-value, but this time the models will be a bit different
Operational definitions
How researchers specifically measure a concept
Independent variable
The variable the researcher manipulates ad controls in an experiment
Dependent variable
The variable the researcher measures but does not manipulate in an experiment
Random assignment
Using a probability-based method to divide a sample into treatment groups. Critical to experimentation because if the only difference between the two groups is the independent variable, we can infer that the independent variable is the cause of any observable difference between the two groups
confounds
Things that could undermine your ability to draw casual inferences
How to prevent confounds
using a double-blind procedure
double blind procedure
neither the participant nor the experimenter knows which condition the participant is in
correlational designs
when scientists passively observe and measure phenomena. We do not intervene and change behaviour as we do in experiments. Identify patterns of a relationship, but usually cannot infer what causes what
Correlation coefficient
Provides information about the direction and strength of the association between two variables.
Positive correlation
The two variables go up or down together
Negative correlation
the two variables move in opposite directions. One goes up, and the other goes down
Strong correlation
The r value will have a high absolute value. If the value is large, it has a strong correlation
Weak correlation
The two variables correspond some of the time, but does not most of the time. R value has a low absolute value
Uncorrelated
if two variables are so weakly related as to be unrelated. R value will be zero or very close to zero
Problem with correlation
Correlation does not mean causation- an often repeated phrase among psychologists
Qualitative Designs
Includes participant observation, case studies, and narrative analysis
case study
Involves an intensive examination of specific individuals or specific contexts
Narrative analysis
Centres around the study of stories and personal accounts of people, groups, or cultures. Researchers examine personal testimonies in order to learn more about the psychology of those individuals or groups
Quasi-Experimental design
An experiment that does not require random assignment to conditions. Relies on existing group memberships, cannot reasonably draw the same conclusions that you would with an experimental design
Longitudinal studies
A study that follows the same group of individuals over time. Provides valuable evidence for testing many theories in psychology, but they can be quite costly to conduct, especially if they follow many people for many years
Surveys
A way of gathering information, using old-fashioned questionnaires or the internet. Can reach a larger number of participants at a much lower cost. Usually used for correlational research
Limitations to correlational and quasi-experimental research
Many practical concerns may influence the decision to use one method over another. Another consideration is selecting a research design is the ethics of the study. Ethical considerations are a crucial factor in determining an appropriate research design
Placebo effect
When receiving special treatment or something new affects human behaviour
Laboratory experiments and their limitations
can clearly separate cause from effect and therefore establish casuality. Clear that a scientific field that is mainly based on controlled laboratory studies ends up lopsided. Accumulates a lot of knowledge on what can happen under carefully isolated and controlled circumstances, but has little to say about what actually does happen under the circumstances that people actually encounter in their daily lives
generalize
In research, the degree to which one can extend conclusions drawn from teh findings of a study to other groups or situations not included in the study
Internal validity
The degree to which a study allows ambiguous casual inferences
external validity
the degree to which a finding generalizes from the specific sample and context of a study to some larger population and broader settings
field studies
allows for the important test of how psychological variables and processes of interest “behave” under real world circumstances
highly sophisticated and carefully controlled
Offer ways to isolate the variety of neural, hormonal, and cellular mechanisms that link psychological variables such as chronic stress to biological outcomes such as immunosuppression
ecological validity
used to refer to the degree to which an effect has been obtained under conditions that are typical for what happens in everyday life
Face validity
the degree to which a procedure or method measures what it intends to measure
Experience sampling method
A methodology where participants report on their momentary thoughts, feelings, and behaviours at different points in time over the course of the day
ecological momentary assessment
an overarching term to describe methodologies that repeatedly sample participants’ real work experiences, behaviour, and physiology in real time
Diary method
A methodology where participants complete a questionnaire about their thoughts, feelings, and behaviour of the day at the end of the day
Experience sampling and related momentary self report
have helped make progress in almost all areas of psychology. Ensure receiving many measurements from many participants, and has further inspired the development of novel statistical methods
Day reconstruction method
A methodology where participants describe their experiences and behaviour of a given day retrospectively upon a systemic reconstruction of the following day. Developed to obtain information about a person’s daily experiences without going through the burden of collecting momentary experience-sampling data
Electronically activated recorder (EAR)
a methodology where participants wear a small portable recorder that intermittently records snippets of ambient sounds around them. Allows the observation to be done in peoples’ everyday lives. Coarse documentation of daily activities and social encounters. Rich data, been used to observe cultural and gender differences in socialbility.
Examples of EAR
time-lapse photography, personal and professional spaces, contents of garbage, video-based methods
Ambulatory assessment
An overarching term to describe methodologies that assess the behaviour, physiology, experience, and environments of humans in naturalistic settings
Linguistic analysis
A quantitative text analysis methodology that automatically extracts grammatical and psychological information from a text by counting word frequencies
