PSY 313 Exam 2 Study Guide Research Ideas Logical Follows facts or observations but not necessary for a good idea Testable All variables can be measured Refutable Falsifiable Can be proven wrong Defining and Measuring Variables Theory An idea about how the world works or an idea about the world based on empirical data An explanation of how scientific laws fit together In Psychology theories describe behavior and underlying causes predict behavior and control manipulate change behavior Hypothesis a specific idea about the relation between constructs derived from Must be testable and refutable Construct Concept of interest that is not directly observable stress attention love Operational Definition Specifies how each construct is measured Turns construct not directly observable into something measurable and observable theory memory Types of Research Descriptive Provides a snapshot of the world not concerned with relationships between variables but rather a description of the variables themselves Correlational Describes how two variables are related does not equal causation Experimental Establishing a cause and effect relationship between two or more variables testing a hypothesis has constructs and independent variables and dependent variables Variables measurable attributes that vary Independent Variables What the experimenter manipulates Dependent Variables What the experimenter measures Potential Problems Reactivity The idea that when you know you are being watched you act differently People modify their natural behavior when they are being watched Demand Characteristics People might do what they feel is expected of them based on clues from the researcher or research design Experimenter Expectancy Effect Experimenter manipulates experiment either consciously or subconsciously to produce the results that they expect Single Blind Study Experimenter does not know hypothesis or the condition the participant is in Double Blind Study Neither experimenter nor participant knows the condition Sampling Research Participants Population The large group of people experimenters sample from ex for college age study the population is all college students Sampling How we select people from the population Representativeness Bias Systematic difference between sample and population Stability Spread or variance of the sample Central Tendency Mean Average Sum the observations divide sum by the total number of observations Best to use in most situations Median Order the observations by magnitude find the middle value Best if there are extreme values ex housing prices Mode The most common observation useful when decimals don t make sense Positively Skewed Tail points towards the positive end median mean Negatively Skewed Tail points towards the negative end mean median Dispersion Variance The average of the squared differences from the mean Standard Deviation A measure of how spread out the numbers are Standard Scores an individual s score expressed as the deviation from the mean score of the group in units of standard deviation X M S When converting to a above or below we use a normally distributed curve Correlation Strength Numerical value close to 1 or 1 Measured with Pearson s R No relationship 0 to 09 Small weak correlation 1 to 29 Medium moderate correlation 0 30 to 49 Large strong correlation 50 to 1 0 Form Pattern in the data usually linear not a curve draw individual line through data and individual scores should cluster around it Direction Positive or negative 3rd variable problem When a 3rd variable correlates with the two variables of interest we don t know which one is the cause Pearson s R correlation coefficient A number that tells you the strength and direction of the correlation Must be between 1 and 1 Scatter Plots 1 point is one observation participant Causation Directionality Causation requires all of the following Correlation Temporal Precedence Ruling out all other 3rd variables Temporal Precedence The cause precedes comes before the effect This can also be proved with experiments not surveys or other forms of correlational research Reliability Is the measurement consistent and stable will it produce the same result over and over Random Error A perfectly sound measure may produce different values ex intrinsic noise weight difference after drinking a liter of water observer error when reading from a scale etc Cancels out because it is random Systematic Error Typically come from measuring instruments either the machine is not measuring correctly or the experimenter is using the machine wrong Inter rater reliability Measures how reliable the experiment is based on the idea that different raters or experimenters should have the same or very similar observations Test retest reliability Measures how reliable the experiment is based on the idea that you should get the same measurements when giving the test at different times Split half reliability Measures how reliable the experiment is based on the idea that you should get the same or similar results when giving someone a test that assess the same construct Validity is the experiment answering the question being asked External Validity How far can the results be generalized to different times places and people Population Other participants cultures genders ages etc Temporal Other periods of time of the day year generations Ecological From lab situation to the real world Construct Validity Does the experiment really measure what the research claims it measures Are the operational definitions reasonable measures of the construct correlated Convergent Validity Shows that two measures of the same construct are Divergent Validity Shows lack of correlation between measure of interest and a different measure Internal Validity Is there anything preventing a causal conclusion Extraneous Variable Any variable that is not a dependent variable or independent variable Confound An extraneous variable that systematically varies with the manipulated variable and explains data Assignment self selection self selection or improper assignment to condition History Uncontrolled events that happen mid experiment Maturation Participant changes over time Instrumentation Change in the ability to use instrumentation or in the measurement device itself Testing effects Change in performance due to practice or fatigue with the material Experimental Control 4 Elements of Experimental Design variable Control eliminate confounds Manipulation create treatment