Psychology Research & Critical Thinking

Conducting Research in Psychology

• Theories and hypotheses guide psychological research.
• Psychologists employ various research methods.
• Psychologists face specific issues when conducting research.

Evaluating Evidence and Critical Thinking

• A good explanation transcends merely naming something; it requires thorough research.
• Psychological researchers utilize scientific methods but encounter challenges unlike those in chemistry or physics.
• One such challenge is sampling, where psychologists must consider the representativeness of their study group.
• Ethical considerations present another challenge, imposing stringent limits on psychological studies, even with animal participants.
• People in research studies sometimes behave differently simply because they know they are being watched.
• General scientific principles and specific adaptations by psychologists will be explored.

The Scientific Approach

• Science originates from the Latin term for "knowledge."
• Science involves a quest for knowledge rooted in carefully observed and replicated data.
• Researchers aim to see reality without bias.
• Using the scientific approach reduces potential biases and errors in data measurement.

Steps in the Scientific Approach

• Perceive the question.
• Form a hypothesis.
• Test the hypothesis.
• Draw conclusions.
• Report the results.

Perceiving the Question

• Research initiates with careful observation.
• Identification of noteworthy occurrences in the environment prompts pursuit of underlying explanations.
• Aggressive behavior in children after watching violent cartoons can be an interesting thing to investigate
• Consider whether cartoon violence contributes to aggressive behavior in children; this inquiry stems from a descriptive goal of understanding the situation.
• After formulating a question, the subsequent phase entails devising a provisional response or elucidation for the observed action, termed a hypothesis.
• It is crucial to distinguish between questions that can be scientifically or empirically studied versus those that cannot.
• The question, "What is the meaning of life?", cannot be addressed using a scientific approach. Empirical inquiries, in contrast, are grounded in direct observation or firsthand experience.
• The question of whether life once existed on Mars is empirical, and can be disproven or proved with real evidence.
• While the meaning of life is a matter of personal belief, scientists need objective evidence to substantiate their claims.
• Psychology researchers seek answers to empirical questions, selecting suitable research methods based on the specific scientific question they aim to address.

Forming a Hypothesis

• Researchers formulate and assess a hypothesis, a precise predictive assertion, to progress beyond mere observations in pursuit of explanations.
• Based on the surroundings, forming an informed prediction concerning the cause for observation and then creating a statement to test is necessary.
• Hypothesis testing is central to scientific investigation and significantly contributes to the development of theories.
• A well-formulated theory should pave the way for the creation of hypotheses based on the theory.
• An “if-then” statement can help in understanding hypothesis formation.
• For instance, the statement, "If the world is round, then a person should be able to sail in a straight line around the world and come back to where he or she started," illustrates this concept.
• The theory part of the statement is grounded on observations and facts that as a ship sails toward the horizon, it gradually disappears.
• The "then" is a testable prediction based on the theory.
• The scientific approach requires a commitment to seeking knowledge even when it challenges pre existing beliefs.
• Hypotheses must be falsifiable, so to test a hypothesis involves attempting to disprove it.
• Hypotheses must also be testable - or able to reveal if it correct or not.
• For the example of cartoon violence, "If exposure to violence leads to increased aggression in children, then children who watch violent cartoons will become more aggressive.” demonstrates what should be tested.
• Creating a hypothesis based on observations serves the purposes description and explanation.

Testing the Hypothesis

• The strategy to validate a hypothesis hinges on the type of response desired.
• Detailed observations, executing surveys on large population samples, or designing deliberate experiments altering one factor are all methods to test a hypothesis.
• A good approach to the cartoon violence is an experiment where children are split into groups and measure behavior.

Method

• Hypotheses can be tested in a variety of ways.
• Each method has different strengths and weaknesses.
• Confidence in a conclusion is increased through studies that all point to the same thing, when tested using different methods.

Testing the Hypothesis (cont.)

• What should you do with the results?
• Testing the hypothesis gives explanation for behavior.

Drawing Conclusions

• Results of the hypothesis testing will show if the data supported initial observations, or that the hypothesis needs to go back to step one and think of a new explanation.
• Hypothesis testing yields data subject to analysis, frequently employing statistical methodologies to arrange and improve the data.
• Making predictions comes from drawing conclusions.

Reporting Results

• Report the investigations success or failure to other researchers.
• Just because a study did not give proof to support a hypothesis does not mean it is wrong.
• There might be some factors that interfered with the study.
• Other researchers need these finding to increase more knowledge abouth the answers to questions.
• The last part of a scientific investigation involves sharing the findings.
• One should write up what you did, why you did it, and what you found.
• The findings are supported better when the study is replicated.
• Behavior can be predicted based on the findings and those findings can be used to control behavior, therefore achieving the last goal of psychology is achieved.
• Replicating studies isn't easy and journals usually embrace new results instead of "old" knowledge.
• Some replicated studies are not as strong as the original ones.
• This can be called a "replication crisis"
• Researchers are responding to this and retesting to provide extra evidence.
• Small affects can be real in teaching, organization and therapy.
• One teaching method may work better than another but researchers cannot copy the advantage in every study.
• Meta analysis handles the trends in data.
• Meta analysis combines results from many studies and analyzes them as though they were all one big study, as well as determines which change in procedure decrease or increase the effects.

Conducting Psychological Research

• The primary goal is to understand research for those hoping to gain research skills.
• Questions should be asked to decide if evidence is valid and what conclusions can be made.

Operational Definitions

• Operational definitions measure something and turn it into numerical value.
• They are not like dictionary definitions.
• They let a person measure something and let the research happen.

Population Samples

• Convenience samples are chosen because of their ease of study.
• Representative samples closely resemble the population.
• These samples give someone equal chance of being selected.
• The larger a random sample, the less likely there will be results that differ from the population.
• Cross cultural samples have people from 2 or more cultures.

Observational Research Methods

• There are many research questions that exist and the usage of these methods depend on what the researcher wants to answer with their questions.
• When you want information about what has happened or what is happening then select a description that gives details.
• This includes:naturalistic observations, laboratory observations, case histories/case studies, and surveys

Naturalistic Observations

• This is all about looking at what happens under normal conditions.
• For example with ADHD children it is difficulty remaining in their seats, don’t pay attention, rudely interrupts other people, angry when they don’t get their way (in school).
• The children at home do not respond when called and throw tantrums when frustrated.
• This forms a list of basic ADHD symptoms.

Disadvantages of Naturalistic Observations

• Values and own beliefs can disrupt cause misinterpret behaviors.
• These observations take time and require effort in training.
• The settings are uncensored, this may make it hard to find causes for these problems.

Advantages of Naturalistic Observations

• It's a way to learn about behaviors in regular places that are impossible to duplicate in labs.
• People being observed would not be motivated to do what the researcher hopes they would, because people in naturalistic observations do not know they are being observed.
• Data will be collected first hand, meaning it is more accurate than recollections.

Laboratory Observations

• It is easier than natural settings to observe behavior of people and animals.
• For example, one might record a babies reflection but it is difficult to set up that kind of setting in a one-way mirror.
• A lab would offer number of infants, their age, and everything else that's going on in a laboratory.

Advantages and Disadvantages of Laboratory Observations

• It can become an artificial space that causes changes in reaction among people or animals, in comparison to the real world. This helps one have control of the observer

Case Histories/Case Studies

• A thorough description of a person is needed, this includes medical condition, disabilities/abilities, life history, experiences, anything relevant.
• It is similar to naturalistic observation, but it focuses on one single person.

Disadvantages of Case Histories

• There can be incomplete and inaccurate recollections
• A researcher's intentions may get in the way while studying the person.
• The individual may be atypical or it may be misinterpreted if the person has pre-ideas in what to look for.

Advantages of Case Histories

• Detailed accounts and information of the life and behaviors that are not understood any other way.
• It provides understanding on the development of psychology problem or for examining patterns of behaviors.
• This is a great way to study rare things.

Surveys

• It's discovering the knowledge of certain actions, beliefs and mindsets.
• This is finding many individuals-e.g. person to person, by the phone, by mail etc. to answer questions about a topic.
• For example: "How many children have ADHD?"

Disadvantages of Surveys

• Many researchers will use readily available sample like college students.
• College students aren't really good samples of the public, being that they are all white and well educated.

Disadvantages of Surveys (cont.)

• Traffic is seen as more of a contributor to air pollution by 24% and industry by 57%, when asked " would you say that industry contributes more or less to air pollution than traffic?" It is switched to 45% traffic and 32% industry with the swapped question, "would you say that traffic contributes more or less to air pollution than traffic?"
• The public will give inaccurate responses from forgetting, lying etc.
• In the black community, it's believed that the problems black faced were self imposed, so using a black interviewer with a 46% agreement versus a white interviewer at 62% agreement goes to show this.

Advantages of Surveys

• Quick and effective instruments for gleaning insight from an expansive subject base.
• Surveys enable identification of issues and assessment treatment programs.

Correlation Studies: Finding Relationships

• The following discussed methods provide details of one's behavior.
• Correlations and experiments are the only two ways more than just an explanation is offered.
• Analyzing statistical information by organizing numerical information so that it is easier to look for patterns in the information is a good way to correlate.

Correlation Studies

• Correlation is a way to measure connection of two or more separate aspects.
• Variables can be measurable, such as age, and gender.
• Controlling any variables allows correlations studies to identify the connection of 2 variables.

Correlational Studies

• Correlation coefficient, indicated with a numerical estimate, determines two aspects in relations to the variables.
• Those are relations and strength.
• Number being closer -1.00 or+1.00, shows greater potential for relations.

Disadvantages of Correlational Research

• A failure to find causes of problems in experiments.

Advantages of Correlational Research

• It assists behavior predictions (if analysts are aware of one variable, they could predict others.

Experiments

• Experiments determines the cause of a behavior.
• An experiment involves manipulative investigations over one variable while measuring other variables.
• The purposeful changing to variable, while holding other variables unchanging is key.
• That way there can be behavior changes and it is known it must be from manipulation.

The Variables

• The manipulated variable is independent.
• The measured experimenter that affects determines how it was affected is dependent.
• The first makes an effect on the latter.
• Confounding is the interference of independent variables.

The Variables (cont.)

• Researchers may discover cartoon violence is related to aggressiveness, but not be certain if that behavior came from the cartoon or the time of day, etc. For this, confounding and variable control is needed.
• Say if violent cartoons cause children from aggressive families to act out, this causes effects that are mixed/confounded.
• As a result, the researcher would not understand is the behavior is because of the cartoon or because the children like aggressive play.

The Variables (cont.)

• Two experiment types that combat confounding variable are seeing the groups of violent cartoon watchers versus non-violent watchers, seeing if their behavior was caused by violence.
• After you would want to understand aggression in both cartoons via measurement.
• If it was significantly bigger in violent cartoons, one could assume cartoons had a big influence on being aggressive.

The Variables (cont.)

• Being an experimental group includes those who've been exposed with a independent entity.
• Control and non-violence should have no effects on the second group of patients.
• In the case there were a large amount of test subjects, assuming violent cartoons will not affect someone isn't good enough.

The Importance of Randomization

• Controlling extraneous variables is achieved by randomly putting the participants in conditions.
• There's equal probability of being assigned to conditions with randomized assignment.
• Having random assignments will make or break experiment qualifications.
• Drawing from a hat ensures each person receives the same chance of getting pulled from the group.

Other Problems

• Some problems can show from doing things on small controlled groups and random placements.
• These problems are common between the people and because their thoughts and pre existing ideas have more of an effect.

Other Problems (cont.)

• Experimenter effects are in the placebo effect. To understand this say we need to test of Alzheimer's drug through samples on the affected patients, while giving one group a drug and the other nothing.
• It necessitates memory testing before and after in the experiment.

Other Problems (cont.)

• But there may be additional problem to the strategy to measure effects. Can they be certain the drug itself made them better or if patients knew this should worked and tried to do it themselves?
• There are things to consider concerning the placebo effect.

Other Problems (cont.)

• Medical research requires control groupings or substitutes which show affected patients.
• An equal chance of a placebo effects shows patients variable success to a placebo rather than a normal one.

Other Problems (cont.)

• Another way to say the findings had problems, when it isn't due to real human reasons, experimenter effect takes charge.
• As it has been talked about in natural observations, watchers can be biased as well.
• Observer biases can bleed into the experiment.
• When a tester measures DVs, body language can be used through tone or just contact of eyes to give away hints.

Other Problems (cont.)

• Regardless if it has been intentional, it's going to happen regardless.
• Imagine an Alzheimers experimental scenario, then to make this all worth it, test patients second memory.
• The testers need to pay attention in the testing to say they got better.
• As an effect to say someone does a better job it helps say something on the response pattern-the action was more effective.

Other Problems (cont.)

• To prevent it a tester can use single or double blinds studies.
• The easiest fix, to avoid fake effects, is to give some controls a test sample and have that affect behavior, if any.
• The samples would need to sugar because they won't make anyone sick or have reactions.
• The single blind in particular has "Blind", new experimental changes are made under these changes.
• Researchers found success in new testing that were successful in testing the standardization while lowering potential (1968).

Other Problems (cont.)

• Actual readings were randomized with different performance measures.
• Attitudes made big changes.
• A double blind experiment has nobody knowing the getting or making to see dependent variables - or is aware of effects.

Disadvantages of Experimental Research

• Participant size is a limit for the experimental process because the double blind can cause new ethical quandaries over studying.
• Certain data cannot be experimented for those issues.

Advantages of Experimental Research

• A controlling variable will likely give data on an extraneous group to find relationships among a random test sample.

Ethical Considerations in Research

• There's basic considerations and ethics with humans.
• Non humans also require ethics tests before experimentation.

Humans

• There needs to be some type, an Ethics Code.
• A committee, and if there is, make clear of if there is deception.
• This, however, requires informed consent.
• Participants need protecting from harm and from the need for honesty.
• All data is private. If there is a data breach a proctor or someone is responsible for taking out or making those consequences be cleaned.

Animals

• Over 20M are used to do biomedical research: like psychology, biology in medicine or drugs.
• Only 7-8% research on rats, mice, and rodents, and rest are all cats and primates.
• With the following, comes issues such as: stress, testing, depression, pain, and etc.
• Animals don't consent to these.

Animals (cont.)

• If humans answer questions about those difficult experiments, animals do as much as those participants.
• Animals do a shorter time while long effects are found.
• Animals are manageable because they can control living and habits, as well as control genetics.
• Animals all show simpler acts - that are more easily effected by experiments.

Animals (cont.)

The list:

1. Animals make more research for what may be deemed harmful or problematic.
2. There is some catch to using animals to save people.
3. It is far by cruel, but without these processes some would never find it.
4. Still, you will hear those thoughts for and against that topic.

Animals (cont.)

• As has show an emphasis of what may hurt a specimen in surgery and it must be done carefully.
• If there is some death to come, the event should be followed with all that needs to happen there.

Measuring and Analyzing Results

• Explain results by the study by needing the results.
• Statistical descriptions helps the data.
• Looking at test scores by checking modes, range, median helps get an view for data.

Descriptive Statistics

• Using averages in data sets to look at number cases and to see information with the normal and bell curves.
• That can be misleading sometimes.
• Medians are better for population when symmetry is off.
• Middles helps in range, while also not needing to change score.

Descriptive Statistics (cont.)

• Most frequent scoring isn't helpful with normal circumstances.
• If it falls as bimodal a distribution by distribution is helped.
• Variation metrics find space around mean - with range and deviation to measure.

Range

• "Highest and Lowest" test results are simple but don't help to read.
• This helps measure amount difference on normal spread of the score.
• Small SD results tend to be clustered near the mean.
• Large SD data results are spread.
• It measures scores on a variety of data.

Evaluating Results: Inferential Statistics

• Researchers want people with random assignments into two separate groups who wish to leave smoking behind with those: with the help of punishment or rewards.
• Most tend to partake in ten before with both. After, punishments tend to get 7.5 in average while the former is at 6.5
• With this data, what can be determined?

Need to Know More Info

• Numbers
• Knowing how much variation occurred
• Figuring out the behaviors were close to average or extreme amounts of activity skew result averages.

We Evaluate the Results with Inferential Statistics

• An inference to sample or to assume things of an average or small set for that type.
• Stats calculate if there's an chance of a difference in what you observed.
• Summary is made (p value) where .05 states they have less than 5% to have randomly produced data.

We Evaluate Results with Inferential Statistics (cont.)

If there happens to be small number results that have big P, chances suggest the events to arise through chance.

• The significance depends on what there is about data of those small or extreme cases of people in each set.
• The amounts with such individuals