Spring 2014 Midterm #2

The flashcards below were created by user deshinan.shs2012 on FreezingBlue Flashcards.

  1. What issue did psychologists face when collecting homicide records of infants by biological vs stepparents when trying to support their hypothesisn "People are less likely to be killed by kin than non-kin?"
    Their hypothesis was not clear in asking number of deaths vs rate of deaths.
  2. Name three things to keep in mind when constructing survey questions.
    1) Break a question into its simplest part 2) Make questions directly relevant to the hypothesis 3) Ask frequency questions on an open-ended scale unless there's a way to easily count the occurrences
  3. halo effect
    Some third person's opinions are in the question, associating feelings with someone in the question.
  4. How do you get rid of the halo effect?
    Get rid of the other person's opinion.
  5. How do you fix a doublebarreled question?
    Break it up into 2 questions
  6. Advantages of open-ended response scale in survey questions (2)
    Unexpected, unrestricted responses
  7. Advantages of close-ended response scale in survey questions (2)
    Easy to code, ensured intended options
  8. Disadvantages of open-ended response system in survey questions (1)
    Must code the answer options in order to survey the answers
  9. Disadvantages of close -ended response system in survey questions (1)
    Putting words into respondent's mouth
  10. Choices on a frequency scale anchors a(n) _____ scale. a) average
    b) linear
    c) correlational
    d) ordinal
    a) average
  11. What caused an uneven response pool between two randomized groups when asked how many hours per day they watched TV?
    A frequency scale with an "anchored" average was used to survey people. Unless an event is small enough to count, an open-ended scale must be used.
  12. Why is it important to make survey answer options mutually exclusive?
    The actual hypothesis must be precise and exhaustive options must be included.
  13. What is a cieling or floor effect due to?
    Insensitivity of a survey response close-ended scale. All the responses are too similarly coded so it shows no difference in the dependent variable.
  14. How may the order of survey questions affect the responses? (2)
    • 1) Assimilation effect: when current mood or feelinh produces a more positive effect target evaluation, or when negative mood or feeling produces a more negative effect target evaluation.
    • 2) Contrast effect: when positive memories create a negative target evaluation, or when negative memories create a positive target evaluation.
  15. Give one example of a social desirability bias.
    When both sexes were asked the number of sexual partners they had intercourse with, men reported a higher number than women. This may be due to a stigma of women being frowned upon for promiscuity.
  16. Gives other words that mean sample. (4)
    1) subgroup 2) strata 3) stratified 4) subset
  17. random sampling
    representative of the population of interest
  18. simple random sampling (3 part def)
    each person of the population has an equal chance of being included; most accurate; probability sampling
  19. population
    group of interest to researcher
  20. sample
    subset of population
  21. sampling frame
    actual individuals from which sample is drawn; probability sampling
  22. stratified random sampling
    the population is first 1) divided into subgroups, then 2) we do random sampling from each subgroup; probability sampling
  23. cluster sampling
    1) groups of individuals are identified 2) clusters are randomly divided 3) all individuals within a selected cluster are sampled; probability sampling
  24. haphazard "convenience" sampling
    whoever is available to test if Variable A causes Variable B; nonprobability sampling
  25. quota sampling
    a sample is chosen to reflect the numerical composition of various subgroups in a population; nonprobability sampling
  26. Is accuracy kept with a sample of 150 people to transfer to an external validity of 15 million people? Explain.
    Yes; because a representative random sample can often estimate the tue population mean very accurately
  27. What does convenience sampling miss?
    It misses questions or external validity
  28. In experimental design of students getting cash rewards for good grades, what was the problem with having a posttest only experiment?
    A correlation doesn't apply because the dependent variable is measured on the results of entire groups.
  29. In a pretest-posttest experiment, what are some added bonuses to finding out the correlation that just a posttest experiment does not? (3)
    It can show if the inital randomized groups are equal, it can measure changes within an individual instead of across a group as a whole, and it can assess the effects of a manipulation.
  30. What are disadvantages of having a pretest ad part of an experimental design?
    It's expensive, and it may affect the posttest results, due to practice.
  31. If the effects of a manipulation at the posttest differ, then the pretest...
    affected the posttest. This is because no other factor could have gone wrong if the experiment is completely randomized and the research assistants acted the same throughout their roles.
  32. repeated measures design (def & ex)
    • Each participant takes part in each experimental condition
    • Ex] Half participants bleed for first 6 months / 1 year and half participants bleed for last 6 months / 1 year
  33. Advantages of repeated measures design (2)
    • Each subject is their own control
    • It's easier to detect the effect of a manipulation with fewer test subjects
  34. Main disadvantage of repeated measures design (1)
    Order effects
  35. Repeated measures example experiment
    Blood letting changes doctor health rating
  36. matching pairs design
    Subjects are matched on a characteristic that is related to the dependent variable, and THEN randomly assigned to conditions
  37. What is the best use of the matching pairs design for an experiment? (2)
    • Small sample size
    • Ensuring initial groups are equal
  38. Avantages of matched pairs design (2)
    • Decreased subject variability and the independent variable is easier to detect because individuals are compared to themselves
    • No order effects
  39. Disadvantages of matched pairs design (3)
    • Costly & risky measurement of the matching variable
    • MUST relate the matching variable to the dependent variable- best when dependent variable IS what you decide to match all the pairs based upon
    • May be unnecessary if sample size is large enough
  40. Name the 3 order effects
    Practice, fatigue and contrast
  41. practice
    With more practice, performance improves
  42. fatigue
    With more time, performance decreases
  43. contrast effects
    • When a positive first event causes the subsequent event to be SUPER negative
    • When a negative first event causes the subsequent event to be SUPER positive
  44. contrast effects example
    Playing country music as the first out of two music samples makes me rate it a 5. Playing country music after playing acoustic indie (10) makes me rate country music a 1.
  45. How do you deal with order effects?
    Counterbalancing- present all of the possible conditions in all of the possible orders
  46. counterbalancing (3)
    • A technique to illustrate order effects
    • Ensures that order effects affect each condition equally
    • It DOES NOT remove order effects: it just makes their effects be seen
  47. Name the 2 expectancy effects
    • Demand characteristics
    • Placebo effect
  48. demand characteristics
    Subjects try "helping" the experimenter achieve the perceived demanded outcome if they think they know or if they are told of the hypothesis
  49. How is a low placebo effect compared to the actual effects in an experiment interpreted?
    If the randomized group who had ACTUAL drugs report a HIGHER success rate than the > placebo drug group, the drug has actual benefit.
  50. Name three experimenter bias solutions.
    • Make the experiementer blind to conditions
    • Run all subjects at the same time and place simultaneously so there is only one interaction for each phase of the experiment per person
    • Play automatic recordings of dependent measures
  51. cross-sectional research
    Subjects of different ages are studied at one point in time
  52. What is the independent and dependent variable of a cross-sesctional research study?
    • IV: age
    • DV: any measurement
  53. Why are the results of crosssectional research studies ambiguous with results open to alternative interpretations?
    Because routinely collected data does not normally describe which variable is the cause and which the effect. Cross-sectional research studies using data originally collected for other purposes are often unable to include data on confounding variables. For example, data only on present alcohol consumption would not allow the role of past alcohol consumption to be explored.
  54. Avantages of cross-sectional research (2)
    • Cheap and fast
    • Common and useful in child development psychology where strong cohort effects wouldn't be relevant
  55. Disadvantages of cross-sectional research (2)
    • Age and cohort effects and confounded
    • Can't differentiate developmental changes from cohort effects
  56. cohort effects
    Being born and raised in a time and place where you and all the other members of your cohort has similar experiences that all other groups don't have, therefore making it an exclusive trait to just your cohort
  57. longitudinal experimental design
    Measures the same persons for dependent variables at different ages in their lifetimes, even until they die
  58. Advantages of longitudinal experimental design (3)
    • Can document within person
    • Identifies age-related changes that cross-sectional research cannot
    • May reveal cohort effects when compared with results from cross-sectional research
  59. Flynn effect
    The absolute scores on IQ tests are increasing in industrialized countries, so intellect is difficult to measure over long spans of time
  60. Limitations of longitudinal experiments (4)
    • Many subjects will die, shifting results
    • Better-educated and healthier humans are likelier to sign up for these kinds of studies
    • Not a randomized study, so it can't be broadly generalized
    • An age-related trend may differ by cohort
  61. sequential experimental design
    Cross-sectional + Longitudinal
  62. Successful manipulation checks include three markers:
    • Checking for accurate and precise internal validity so that ONLY the independent variable is being manipulated
    • A survey at the end of the study to not give the hypothesis away and produce expectancy effects
    • Elimination of all alternative explanations
  63. straightforward manipulation
    Changing the independent variable condition and measuring their effects
  64. staged manipulation
    A forged movie involving deception and confederates, research assistants playing roles in the study
  65. Asch's 1941 study of conformity
    • 80% were wrong because they conformed at least once
    • 2% were wrong in private testing
  66. naturalistic observations
    • analyze observations and form hypothesis post hoc: after the fact, just making observations
    • goal is to provide a complete and accurate picture instead of testing hypothesis formed prior to the study
    • Usually qualitative research
    • Often animal research, participant/non-participant observations (Did you join the group to watch from the inside?), concealment/non-concealment observation (Subjects may act differently if the tester is undercover.)
  67. habituation
    when you stay for so long they forget you're there
  68. What is an issue in how to categorize events in naturalistic observations?
    You must create operational definitions. Without operational definitions, how do you ensure observations are objective? Can be useful for scientific research to generate hypotheses and then later go back with scientific research.
  69. In the Robert Cialdini students showing up to a 7 am psych experiment, what technique was used and what were the results?
    • When students were given full disclosure of a 7 am voluntary psych experiment, only twenty-four percent agreed. 
    • When students were told of a voluntary experiment, fifty-six percent agreed, and when later told it was at 7 am, none backed out.
  70. low ball technique
    making an initial clause which the participant will agree to, then adding an extra cost whose negative effects are downplayed
  71. systematic observations
    • only interested in specific behavior
    • quantifiable
    • pre-formed hypothesis
    • hypothesis-driven
    • need reliable coding system (What segment of time to record, operational definitions, etc.)
    • need recording equipment
    • Limitations: reactivity and results depend on the methods and coding
  72. In baboon greetings of diddling male genitalia, what is a signal of its reliability?
    A) the riskiness of damaging the genitalia
    B) the cost of diddling the genitalia
    C) hard to fake the greeting
    D) grooming the genitaliae) amount of time spent
    A) the riskiness of damaging the genitalia
    (this multiple choice question has been scrambled)
  73. costly signaling theory
    signals can be regarded as reliable if they are more costly, risky, and hard to fake
  74. Why are male peacock feathers a reliable indicator of mating quality?
    Peacock feathers are difficult to fake and genetically costly.
  75. Which is NOT considered to be a code in the coding system of a systematic observational study?
    A) a 90 degree head turn while kissing is ride-side bias
    B) brain size is defined as "ratio of brain matter to animal size"
    C) grooming is defined as "one animal manipulated the skin or fur of another"
    D) a measurement of two birds 1.5 meters away is defined as "interaction"
    B) brain size is defined as "ratio of brain matter to animal size"
    (this multiple choice question has been scrambled)
  76. case studies
    • writings, usually qualitative, that provide detailed descriptions of the behavior of one individual, usually in rare circumstances
    • Limitations: Because conclusions are drawn post hoc, causality is difficult to determine
  77. archival research
    • quantitative, hypothesis-driven, mining for data collected by others
    • 3 types of data: 1) statistical records 2) survey archives 3) written records
Card Set
Spring 2014 Midterm #2
Show Answers