Methods of Acquiring Knowledge

Methods of Acquiring Knowledge

• Tenacity involves accepting knowledge based on habit or superstition.
• A disadvantage of tenacity is the information might not be accurate.
• Intuition involves accepting information based on a hunch.
• A disadvantage of intuition is there is no method for separating accurate from inaccurate knowledge.
• Authority involves relying on an authority figure to obtain knowledge.
  • A disadvantage of authority is that authority figures can be subjective.
• Rational method involves seeking answers using logical reasoning.
  • A limitation of the rational method is that people are not always good at logical reasoning.
• The empirical method involves using observation or direct sensory experience to obtain knowledge.
• A disadvantage of the empirical method is observations can be misinterpreted.

The Scientific Method

• Observation involves observing behaviors or other phenomena.
• Forming a hypothesis involves forming a tentative answer or explanation.
• Using the hypothesis to generate a prediction involves applying the hypothesis to a specific, observable situation.
• Testing the prediction by making systematic observations involves carefully and objectively evaluating the prediction.
• Using the observations to support or refute the hypothesis involves comparing the actual observations with the predictions that were made.
• Other elements of the scientific method include that science is empirical, public, and objective.
• Pseudoscience lacks one or more components of the scientific method.

The Research Process

• Qualitative research is often exploratory and uses less structured research methods.
• Quantitative research is generally used to examine relationships between variables through the use of statistical analyses.
• The research process follows a series of steps.
• Step 1 involves finding a research idea by selecting a topic and searching the literature to find an unanswered question.
• The question should be answered by collecting and statistically analyzing sample data.
• Step 2 involves forming a hypothesis.
• Step 3 involves determining how the researcher will define and measure variables.
• Step 4 involves identifying the participants or subjects for the study, deciding how they will be selected, and planning for their ethical treatment.
• Step 5 involves selecting a research strategy.
• Step 6 involves selecting a research design.
• Step 7 involves conducting the study.
• Step 8 involves evaluating the data.
• Step 9 involves reporting the results
• Step 10 involves redefining or reformulating the research idea.

Research Strategies

• Descriptive research strategy involves describing individual variables instead of relationships between variables.
• Correlational strategy involves examining relationships between variables.
• Experimental research strategy involves demonstrating a cause-and-effect relationship between two variables.
• Quasi-experimental strategy involves attempting to demonstrate a cause-and-effect relationship between two variables.
• Non-experimental research does not use the rigor and control that are found in experimental research.
• Population is an example of non-experimental research, and is the entire group of interest.
• Sample is a subset of individuals selected from the population.
• Representative samples have the same characteristics as the population.
• Sample size involves how many participants the study requires.
• Sampling frame is the list of individuals from which a sample is selected.
• Sampling involves selecting individuals to participate in research.
• Sampling process involves selecting a representative group from the population.

Probability Sampling

• Probability sampling increases the chance of selecting a representative sample.
• Simple random sampling requires that each individual in the population has an equal chance of being selected and each selection is independent of the others.
  • Some criticisms of simple random sampling are that it might not select a perfectly representative sample and it requires that the researcher know each individual in the population.
• Systematic sampling involves selecting every nth participant from a list.
• Stratified random sampling involves dividing the population into subgroups and then randomly sampling from each subgroup.
• A benefit of stratified random sampling is that it guarantees that each subgroup will be represented in the sample.
• A disadvantage of stratified random sampling is that the proportions in the sample must match the proportions in the population.
• Proportionate stratified random sampling involves ensuring that the sample proportions match the population proportions.
• Cluster sampling involves randomly selecting clusters of participants from the population.
  • Some advantages of cluster sampling are that it is easy to implement and it is useful when the population is large and spread out.
  • Some disadvantages of cluster sampling are that the clusters may not be representative of the population and there is a risk of selecting a biased sample.
• Multistage sampling involves combining two or more sampling techniques to select a sample.
  • Some advantages of multistage sampling are that it allows researchers to tailor the sampling strategy to their research question and it can save time and money.
  • A disadvantage of multistage sampling is that it requires a thorough understanding of each sampling method.
• Matched random sampling involves matching participants on a variable that is relevant to the research question.

Non-probability Sampling

• Non-probability sampling does not require that each individual in the population has an equal chance of being selected.
• Convenience sampling involves selecting participants who are easy to access.
• Quota sampling involves selecting participants until a certain quota is met.
• Purposive sampling involves selecting participants who meet specific criteria.
• Snowball sampling involves asking participants to refer other participants to the study.
• Volunteer sampling involves selecting participants who volunteer to participate in the study.

Data Collection

• Observation is a common method of data collection.
• Participant observation involves the researcher participating in activities of the people being observed.
• Non-participant observation involves the researcher observing the people without participating in their activities.
• Structured observation involves using a predetermined system for recording observations.
  • This includes event coding, which involves recording each time a specific event occurs.
  • Plus Interval coding, which involves dividing the observation period into intervals and recording whether a specific event occurs during each interval.
• Reliability of observational techniques involves the consistency of the observations across different observers.
• Controlled observation involves observing behavior in a controlled setting.
• Naturalistic observation involves observing behavior in a natural setting.
• Reactivity effects occur when participants modify their behavior because they are aware that they are being observed.
• Full participant observes and interacts but does not inform the subjects of the research.
• Participant as observer observes and interacts with participants, being open about research.
• Observer as participant observes and interacts minimally with participants.
• Full observer observes without interference.
• Strengths of participant observation include gaining insight into the group's culture and activities.
• Difficulties with participant observation include the researcher's presence influencing behaviors.
• Case studies involves studying a single individual in great detail.

The Two Problems

• Causation and the third-variable problem address that a third, unmeasured variable might be responsible for the relationship between two variables.
• Causation and the directionality problem address the possibility that one variable is the cause, but the other variable is the effect.

Conditions and Variables

• Conditions are the values of the independent variable.
• Manipulation involves the researcher changing the value of one variable to create a set of two or more conditions.
• Manipulation and the directionality problem involve determining which variable is the cause and which is the effect.
• Manipulation and the third-variable problem involve controlling for extraneous variables that could influence the relationship between the independent and dependent variables.
• Control involves influencing and holding steady particular aspects of the testing situation.
• Control and the third-variable problem involve controlling for extraneous variables that could influence the relationship between the independent and dependent variables.
• Extraneous variables are uncontrolled variables that can influence the outcome of a study.
• Confounding variables are variables that are related to both the independent and dependent variables.
• Controlling extraneous variables involves holding a variable constant.
• Matching values across treatment conditions and control by randomization can also control extraneous variables.
• Control groups do not receive the treatment, such as no-treatment control groups and placebo control groups.
• Error variance is the variability in the dependent variable that is due to factors other than the independent variable.
• Sources of error variance include individual differences, environmental factors, and measurement error.
• Handling error variance includes reducing error variance and increasing the effectiveness of the independent variable.
• Reducing error variance can be done by standardizing procedures, using reliable measures, and controlling extraneous variables.
• Increasing the effectiveness of the independent variable can be done by increasing the strength of the manipulation.
• Randomizing error variances across groups assumes error variance will be approximately the same from 1 group to another.

Between-subjects Experimental Design

• Between-subjects experimental design involves comparing two or more different groups of participants.
• Characteristics of between-subject designs include each participant being assigned to only one condition.
• Advantages include each score being individual and obtained independently.
• Disadvantages include a large sample size needed and individual differences.
• Individual differences can threaten internal validity.
• Limiting confounding by individual differences includes random assignment to distribute participant characteristics evenly.
• Matching groups is when members of one group are matched with members of another group.
• Restricting range of variability and threats to internal validity can occur.
• Differential attrition occurs when rates of withdrawal from a study are not the same across different groups.
• Communication between groups occurs when participants from one group talk to participants from another group.
• Statistical analyses include single-factor two-group design and single-factor multiple-group design.

Within-subjects Design

• Within-subjects design involves comparing two or more scores for one group of participants.
• Characteristics of within-subjects design include the same participants being used for all conditions.
• Advantages include needing relatively few participants and eliminating individual differences.
• Disadvantages include threats to internal validity when more than one condition is administered.
• Threats to internal validity must be considered.
• Environment variables can threaten internal validity in within-subjects designs.
• History can threaten internal validity in within-subjects designs due to outside events.
• Maturation can threaten internal validity in within-subjects designs due to changes within the participants.
• Instrumentation can threaten internal validity in within-subjects designs due to changes in the measuring instrument.
• Order effects (practice, fatigue, and carry-over effects) can threaten internal validity in within-subjects designs.
• Countering threats to validity involves counterbalancing.
• Matched-subjects design involves matching participants on relevant variables.
• Statistical analysis is needed.
• This involves two-treatment designs and multiple-treatment designs.
• Nonexperimental and Quasi-experimental Research Strategies are other strategies that can be utilzied.
• Nonexperimental research studies do not use the rigor and control that are found in experimental research.
• Quasi-experimental research strategy attempts to demonstrate a cause-and-effect relationship between two variables.
• The structure of nonexperimental and quasi-experimental designs include different types of designs.
• These include:
  • Nonexperimental designs with nonequivalent group designs.
  • Differential research design.
  • The posttest-only nonquivalent control group design.
  • A quasi-experimental design with nonequivalent groups.
  • The pretest–posttest nonequivalent control group design.
  • Nonexperimental pre-post design.
  • The pretest-posttest design.
  • The time-series design.
  • The cross-sectional developmental research design contains, strengths, and weaknesses.
  • The longitudinal developmental research design includes strenghts.