Experimental Notes

Ideas

• birth order and independence

• birth order and academic anxiety/achievement

• beliefs about misinformation in college aged students

• birth order and personality type

• persoanlity trait and persuasion

  • show an ad of some type and see how likely people are to buy the product based on personality traits

• personality traits and emotional expression HEXACO

• family dysfunction and music taste?

• religiosity and what is believed as morality

• Tipping habits

• implicit bias (https://implicit.harvard.edu/implicit/selectatest.html)

• altruism

Scientific Process

• 1. Find an idea

     • read scientific literature

     • personal observations

     • talking to other people

  2. Literature Search

     • read literature

     • cross reference ideas

     • critically evaluate articles

     • discuss with others

       • where does this idea fit in with psychology?

  3. Formulate a research question

     • consider the scientific merits of the topic

     • consider constructs you measure

       • can you measure scientifically?

  4. Select methodological approach

     • descriptive

     • experimental

       • allows us to determine cause/effect relationships

     • correlational

       • allows us to investigate the relationship between two or more variables

       • cannot determine cause/effect relationship

  5. Practical Considerations

     • sample - do I have one/can I get one?

     • cost

     • do testing instruments exist? can I get them?

     • time

     • ethical considerations

  6. Get IRB approval

     • Institutional Review Board

     • never collect data without IRB approval

  7. Is it practical?

  8. Conduct Research

  9. Obtain Results/Data

     • analyze data

     • organize results

     • report results

Finding an Idea

• personal strategies

  • reading novels

  • listening to the radio

  • watching TV

  • these sources of information reveal problems that many need research to aid in their resolution

• “why” questions

• deviations

  • find the exception and evaluate

  • when the general trend itself is odd

• introspection

  • analyze your own behavior

• sustained, deliberate observation

  • case studies

  • participant observation

• definitions

  • define a topic better

• pushing plausible to the extreme

  • push reasonable hypothesis to an implausible extreme

    • ex. extreme eye contact

• interpersonal strategies

  • ask others about ideas for research

  • listen to lectures

  • topics of interest without specific questions

• printed sources

  • journals, text books, technical reports, thesis/dissertations

Proposal Paper

1. title page

   • concise description of the topic

   • APA 7th, student paper protocol

2. abstract

   • 500 word summary of paper

   • write after the paper

3. introduction

   • state hypothesis/research question

   • give rationale for research question

4. literature review

   • background information on topic

   • explain how your idea fits into past research

5. method

   • what are you doing in the study

     • participants

     • measure variables

6. results

   • straightforward

   • analyze the data

   • statistics

7. discussion

   • discuss results

     • was the hypothesis supported?

     • why or why not?

8. conclusion

   • tie report together

   • areas for improvement

   • ideas for future study

9. appendix

   • charts, photos, surveys

Ethical Research

• pseudoscience - activities and beliefs that are claimed to be scientific but are not

  • biorhythms?

• ethical dilemmas

  • does the potential gain in knowledge from the research study outweigh the cost to the research participant

  • Institutional Review Boards

  • APA Code of Ethics

    • investigator’s responsibility to evaluate their study in light of:

      • degree of risk

      • except in minimal risk situations, the investigator has the ethical responsibility to inform the research participants about all aspects of the research

      • if deception or concealment is needed, the investigator must determine if alternative procedures are available that do not employ deception or concealment

        • deception is okay when:

          • the research objective is of great importance but cannot be achieved without using deception

          • on being fully informed later, participants are expected to find the procedures reasonable and to suffer no loss of confidence

          • research participants are allowed to withdraw from the study at any time

          • investigators take full responsibility for detecting and removing stressful after-effects of the experience

      • you cannot coerce a person to participate in research

      • must protect participant from physical and mental discomfort, harm, and danger

      • after the data is collected, the investigator provides the participant with information about the nature of the study and attempts to remove any misconception that may have arisen

      • informed consent - the principle that research participant should be fully informed about all aspects of the study

        • not necessary in some studies (ex. archival research)

      • deception

        • active deception - refers to deception by commission where the experimenter deliberately misleads teh research participants such as giving false information about the purpose of the experiment

        • passive deception - deception by omission where certain information is withheld from the participants

      • debriefing

        • very important to use after deception

        • milgram reported that only 1.3% of his participants reported negative feelings about his experiment

          • 4% believed that the experiment should not continue

        • two goals: dehoaxing and desensitizing

Experimental Research

• experimental research

• independent/dependent variable

• single blind study - subjects do not know whether they are in an experimental or control group

• double blind study - neither subjects or researcher know which subjects are experimental or control

  • both receive placebo

    • placebo - inactive substance or fake treatment used with the control group in a blind or double blind study

• all medications approved by the FDA have numerous double blind studies

• randomization

  • has to happen in experimental design

  • each participant has an equal chance to be placed in either the experimental or control group

    • helps control confounding variables

  • quasi-experimental design is not randomized, still provides valid data

• confounding variables

  • variable that affects the dependent variable so that any change in behavior could be due either to the independent variable or the confounding variable

  • unplanned variables that affect the dependent variable but which are not part of the independent variable\

• operational definitions/operationalizations

  • definition of the concept by specifying the operations required to measure the concept

  • can apply to dependent or independent

  • define to measure

• validity

  • internal validity - says that the experiment results are valid

  • external validity - says that you can generalize the results to other groups

  • usually compete with one another

    • to get high IV, you need control over the situation → control makes the situation less natural which lowers EV

    • internal is the most important

• pre-experimental designs (pilot studies)

  • problems: lack randomization, do not control confounding variables

  • useful: test levels and procedures to be used in the final experiment

  • can be used as a legitimate experiment, but problems should be discussed in report

  • using a large number of participants offsset problem

other types of research

• observational:

• usually correlational so does not involve manipulation of variables

• behavioral categories often used

  • looking for the number of occurrences of behavior

• categories must be operationally defined

  • behavior that is not well defined leads to recording errors

  • results that are difficult to interpret

• to develop defined categories, begin with clear idea about the goals of your study

• quantifying behavior

  • frequency method - you record the number of times a particular behavior occurs within a time period

  • durative method - record how long a behavior lasts

  • intervals method - divide observation into discrete time intervals and then record whether a behavior occurs within each interval

• bias

  • observer bias - occurs when your observers know the goals of the study or the hypothesis

  • interpretation bias - observers sometimes interpret what they see rather than simply record behavior

• naturalistic observation

  • observing subjects in either natural environment without making nay attempt to control or manipulate variables

  • very difficult

    • simply observing may disturb behavior

      • make unobtrusive observations

      • gradually introduce yourself to the environment of subjects

      • abandon direct observation for indirect

• ethnography

  • become immersed in the behavior or social system being studied

    • joining cults, tribes, etc.

• case history

  • descriptive technique where you observe or report on cases

• archival research

  • involves studying existing records

  • historical accounts of events, census data, court records, crime reports

  • have specific research questions in mind

  • consider need to gain access

    • you may have to ask for permission

  • sources: census

• content analysis

  • used when you want to analyze written or spoken record for the occurence of a specific category or event

• one-shot study

  • two variables - X, O

    • X can be a variable that you introduce or one that occurs naturally

  • inferences from the study are based on general expectation of what would have happened if X had not occurred

  • survey also fits this category

• one group pre-test/post-test design

  • O1 (observation 1) X (treatment) O2 (observation 2)

  • ex, test the creativity of a group uof fifth graders (o1) and then the creativity of the group after (o2) they have gone through puberty (X)

operationalization (operational definitions)

• process of defining variables into measurable factors

  • ex. hypothesis: expression of anger builds more anger

  • procedure for identifying instances of the critical terms: increased heart rate, change in tone of voice, physical aggression/violence, scores on anger survey

• define variable to make them measurable

  • define exact measuring method used and allows other scientists to follow exactly the same methodology

• fuzzy concepts

  • vague ideas/concepts that lack clarity or are only partially true

  • important to define the variables to facilitate accurate replication fo the research process

• advantages

  • good for social science determines how researchers measure emotions

  • allows others to replicate research, perform statistical analysis

  • specificity

• disadvantages

  • involves narrowing down from broad, less clear concepts

  • provoke disagreement

  • can be complained as “too narrowly interpreted”

  • disagreements are settled by philosophical, political, pragmatic, moral argument

literature review

• why:

  • science is an organized body of knowledge, not a random collection of facts

  • a finished study answers questions “this is what happened” and “this is where it dits in”

    • to know where it fits in you have to know what was done before

    • literature search

    • background search

• sources:

  • textbooks

    • introductory textbooks provide a broad overview of field

    • time lags, can be biased

    • good starting point

  • monographs

    • book-length sources that go beyond the textbook

    • goes into specific topic areas

    • communicates results of original research

    • makes unique contributions to the field

    • ex. Variety of Religious Experience by William James

  • review sources

    • psychological bulletin: publish evaluative and integrative reviews of substantial research

    • annual review of psychology

      • reviews present selective/evaluative status and recent progress in main areas of the field

      • contains about 20 chapters

      • each chapter contains an extensive reference list

  • original journals

    • most published research appears as articles in journals

    • primary sources

    • present the basic results as interpreted by the experimenter rather than a third party’s compilation or summation

  • psychological abstracts

    • an abstract/short summary of all psychological research including government documents

  • dissertations and thesis

    • most doctoral programs require a thesis/dissertation

    • an original piece of research that demonstrates the researchers knowledge in that area

      • extensive literature review

  • conference papers

    • presentations or conferences can include reports of research, symposia discussions, and invited addresses

    • earliest public report of recent research

    • many conference papers never appear in print

    • attain by:

      • go to conference, take notes, handout

      • email and ask for copy

  • government documents

    • agencies issue thousands of publications each year

    • US government printing office

    • available on internet

  • internet

    • websites dedicated to all kinds of topics

    • use with caution

    • guidelines:

      • authority - is the source reliable

      • accuracy - can the info be verified

      • objectivity - is the info free of bias

      • currency - available and up to date

      • coverage - is the info uniquely available on this site?

  • PsychINFO

    • produced by APA

    • 1900 journals for psychological research

    • indexes books and book chapters dissertations and research reports

    • some entries date to 1880s

  • PsyLIT

    • everything published in psychology since 1974

  • ERIC

    • contains primarily education documents

  • PubMED

• treeing backwards: find the most recent article that deals with topic of interest, then find articles at the reference list at the end

communicating science

• APA recognize importance of communication in Guideline for the Undergraduate Psychology Major by making communication one of the main goals for psych majors

• know your audience

  • several different purposes

  • two distinct audiences

    • professional peers/fellow scientists - need to know what you found and how you found it

    • general public - non-scientists, family members, neighbors, etc, who ay be able to use psych findings to improve their lives

• say it succinctly

  • only long enough to convey the point

  • long-winded explanations make the science harder to understand

  • every sentence has a purpose

• choose verb tense wisely

  • first - what you think will happen

  • next - describe how you did it

  • finish- explain what happened

• choose words wisely

  • specific and precise

  • avoid saying what the study was intended to find

  • never appropriate to say a study can prove anything

  • refer to self as researcher

• keep data clear and organized

  • fact based case to justify your study

  • logical and intuitive organization of ideas

  • taking notes on articles

    • read article at least twice

    • highlight parts of sentences

    • notes in margins

    • transfer notes into doc that organizes and summarizes notes

    • keep notes short and succinct

      • citation, general topic, subtopic ,key research cited in article, overall thoughts

    • for empirical articles:

      • hypothesis/research questions, participants, design elements, variables, for review articles/chapters

    • qualitative article:

      • key theories/ideas (summary of central ideas)

hypothesis

• null hypothesis is tested

• alternative hypothesis (Ha):

  • independent variable caused the result

  • no way to definitively prove (why we test the null)

    • prove that null hypothesis is probably wrong

  • ex. anti-depressants cure depression

  • statistical significance cut off is 0.05 (5% or less)

    • reject the null

• null hypothesis (Ho):

  • all other possible outcomes of study

  • hypothesis of equality

  • ex. anti-depressant increases depression/stay the same

• researcher’s steps for hypothesis testing

  • state alternative and null hypotheses

  • take population to be sampled

  • assume the null hypothesis is correct

  • apply the treatment conditions to randomly selected samples

  • find the significant difference or relationship

  • accept or reject null hypothesis

• reasons to test the null hypothesis

  • it is the most likely explanation for what has occurred

  • no way to verify alternative to the null hypothesis directly

• Type 1 and Type 2 errors

  • type 1 error: if we reject the null hypothesis when it really is true

    • no relationship but claim there is

    • false positives

    • “if you find results and there are none, you have error type 1”

  • type 2 error:

    • false negatives

    • “if you find nothing but results are true, you have error type 2”

• issues with hypothesis generating

  • hypothesis that is too vague to test

    • ex. Ha - love is better than hate

  • wanting to know everything

    • ex. I want to know why people gossip

  • Insisting on making a world changing impact

  • insisting on enjoyment

choosing the right statistical tool

• what’s your plan

  • describe, make inferences?

• what type of measurement are you using?

• how often do we measure each participant?

• how many comparison groups does the study have?

• definitions:

  • measurement scale - classification system used to measure a variable

  • nominal scale - scale where numbers are used as category

  • dummy coding - process of assigning number to represent categories when measuring nominal scale variable

  • dichotomous variables - nominal scale variable with only two categories

  • ordinal scale - measurement scale where numbers are used to rank the variable on some dimension

  • interval scale - scale where numbers indicate an ordering to the measurements and the difference between each measurement value is the same

  • ratio scale - measurement scale where numbers indicate an ordering to the measurements, the difference between each measurement value is the same and there is a true zero point

• descriptive stats

  • frequency of form

    • histogram - type of chart used to graph in continuous variables; the frequency of the measurement is represented with bars that touch to indicate that continuous variable is being graphically displayed

    • frequency polygon - chart used to graph continuous variables; the frequency of a measurement is represented with a point in the graph and these points are connected with a line similar to the histogram

    • frequency distribution table

    • bar chart

    • normal distribution - distribution of scores that resembles a bell-shaped graph

  • central tendency (mean, median, mode)

    • helpful to identify the average score

    • mean → interval and ratio

    • median → ordinal data

  • variability

    • range - measure of variability in the date computed by subtracting the highest score from the lowest score

  • measure associations between variables

    • pearson product moment correlation coefficient (pearson r) - the correlation coefficient used when both variables are continuous

      • tells strength and direction of two variables

      • scale of 0 to 1 or 0 to -1

      • look at absolute value (0.80 is just as strong as -.80)

      • positive correlation - variables move in same direction

      • negative correlation - variables move in opposite directions

• inferential stats

  • samples to make estimates about populations

    • t-tests

    • anovas

conducting interviews

• general guidelines:

  • state purpose of the interview

  • identify potential participants

  • determine questions to ask/question order

• interviewer bias

  • any way that the interviewer influences the participants responses

  • consider demographic characteristics

• qualities of good interviewer

  • prepared

  • fully knowledgable about interviewee

  • proactive questions (clear, don’t sound rushed or nervous)

  • attentive, focused, sensitive, tolerant, appropriate segues, highly organized

• interview schedule

  • type of protocol that includes the questions to ask and anticipated order in which the interviewer should ask them

• remain neutral and avoid any reacts that influence response

• plan for potential follow-up questions

• pay attention to sequence of questions

• conducting the interview

  • establish rapport early

  • interview should feel like a conversation

  • take handwritten notes when possible

  • stick to the interview schedule

  • engage in active listening

surveys

• concise, preplanned set of questions designed to yield specific information to meet a particular need for information about a pertinent topic

• advantages:

  • inexpensive

  • quick

  • uniformity of questions - each respondent sends the same set of questions phrased in exactly the same way

  • standardization

• disadvantages

  • sampling problems

    • is data accurate

    • 1936 literary digest poll

• polling

  • use tracking polls

  • census 0 measure each person about whom w wish to know

  • sampling - selection of people to participate in a research project

  • population - the entire group of people that the researcher desires to learn about

  • sample- smaller group of people who actually participate in the research

  • representative sample - one that is approximately the same as the population in every important research

  • probability sampling - procedures used to ensure that each person in the population has a known chance of being selected to be part of the sample

• types of sampling designs

  • simple random sampling Srs

    • back of the stats textbook?

• systematic sampling

  • using sampling interval or sampling ratio

  • get as generalized sample as possible

• stratifies sampling procedure

  • stratify by subgroups

    • ex. freshman, sophomore, junior, senior

• probability proportionate to size (PPS) sampling

  • in the first stage of sampling, each cluster is given an chance of selection proportionate to its size

• methods of surveying

  • direct (face to face)

    • ex. going door to door, stopping people on street

    • controlled population (students in a class)

  • indirect (ex. mail)

• factors affecting the percentage of returned questionnaire

  • length, reputation of sponsoring agency, complexity of questions, etc.

• in cases where probability sampling is impossible, nonprobability must be used

  • snowball sampling - one individual is contacted, they contact their friends

    • difficulties

  • convenience samples - samples in which the researcher has sampled whatever individuals were readily available without any attempt to make the sample representative of a population

• designing a survey

  • objective - state why the survey is necessary/what you want to find

  • sample - state which population you are interested in

  • questionnaire

    • decide how sample will be surveyed

    • decide on type of questions

  • decide type of nominal scales

    • categorical

      • no real number value (like number labels in a race)

    • ordinal

      • indicate rank on a survey item

      • difference among responses, but does not indicate magnitude among differences

      • Likert scales

        • responses are not bivalent

  • discrete visual analog scales

    • helpful 1-10 not helpful

    • more generic than likert

  • open and close-ended questions

    • may leave out answers forcing subject to choose between inadequate answers

    • working may make choice more appealing

    • guidelines for close ended

      • response categories should be exhaustive

      • answer categories should be mutually exclusive

    • open and close-ended questions can be used together

    • contingency questions

      • one question that determines which question is answered next

  • matrix questions

    • several questions in a row that have the same set of answer categories

    • always include neutral point (undecided)

    • items must be clear

    • categories must not overlap

  • avoid double-barreled questions

    • ex. do you believe in government bailout and no strings?

  • ensure respondent’s competency to answer

  • use short items

    • only long enough to get essential data

  • avoid negative items (ex. is your auto not insured)

  • avoid biased items/terms

• advantages of a survey

  • helps with expenses

  • uniformity of questions

  • standardization

• sensitive questions

  • avoid personally embarrassing questions

  • could admit illegal behavior

  • ensure confidentiality

• format

  • spread out

  • uncluttered

  • maximize white space

• ordering questions

  • ask open-ended questions first

  • don’t randomize, should be flow to questions

  • in self-administered - begin with most interesting set of questions

• instructions

  • clear instructions and introductory comments

  • short intro to make sense of questionnaire

  • how to indicate answers

  • guide about one-word, sentence, or paragraph answers for open-ended

qualitative research

• mixed research methods

  • generic term representing a variety of methods that focus on obtaining an in-depth account of participants perspective of their own world and their experience of events

• mixed research methods

  • blend of qualitative and quantitative methods that capitalizes on the strengths of each to examine a research question from multiple perspective

• qualitative vs. quantitative

  • in-depth info from fewer participants // focused piece of info from more participants

  • values participant perspective // less emphasis on subjective interpretation

  • favors studying participants in natural settings // favors studying participants in controlled lab setting

  • detailed descriptions of bx written or verbal // relies on numbers to describe data

  • no hypothesis testing // almost always test a hypothesis

  • does not seek to establish cause and effect (describing relationships) // sometimes establishes cause and effect

  • very little structure, participant designed // seeks to generalize a study to broader populations

  • avoid summarizing, categorizing, or reducing data // use data summaries, makes generalizations of groups of people rather than focusing in on groups of people

  • research has more influence on data collection // researcher works to minimize data

• similarities with self-report study

  • amount of time and energy

  • relies on small sample which makes it hard to generalize

  • possible that participant may lie/provide inaccurate info

• common terms

  • triangulation - using multiple methods to assess the same information

  • reflexivity - researcher monitors and records their role in the data collection on a continuous basis during the study, which allows for a more accurate assessment of the researchers influence

  • bottom-up approach - researcher develops a theory by exploring a topic using information provided from the participate to direct experience

  • top-down approach - deductive approach, researcher test preconceptions and previously established theories with collective data

  • situated analysis - approach where the researcher examines a topic while it is embedded within its naturally occurring context

  • holistic analysis - researcher examines how numerous properties contribute to patterns within the larger and more complex system

  • phenomenological approach - seeks to understand human experience and the meaning of experiences based on how those involved view that situation

types of qualitative studies

• action research

  • explicitly involves participants in the research and tries to change some aspect of the research’s focus

• case study

  • comprehensive description of a specific organization, group, or person over a period of time that contains information from a variety of sources

• conversation analysis

  • examines patterns of dialogue

  • focuses on turn taking, gaze direction, speaker’s sequence of speech

• content analysis

  • systematic analysis of written work

• ethnography

  • uses detailed long-term observations to place a phenomenon in the cultural context

• focus group

  • data collection format where several participants, likely strangers, are brought together to talk about a specific topic

• grounded theory technique

  • approach where researcher does not have any explicit theories or hypothesis to test prior to research

  • instead uses information from participants to generate the categories and built a theory

• interview

  • mimic a conversation where the research elicits self-report data directly from the participant

• narrative analysis

  • examination of first person stories or descriptions of one’s life

  • analyzed by the researcher from the storytellers point of view

• post-modern approach

  • perspective that questions assumptions about nature and research

  • truth is unique to individuals so we cannot know absolute truth

  • scientific thinking is outdated