Research Methods an der University Of Amsterdam | Karteikarten & Zusammenfassungen

Lernmaterialien für Research methods an der University of Amsterdam

Greife auf kostenlose Karteikarten, Zusammenfassungen, Übungsaufgaben und Altklausuren für deinen Research methods Kurs an der University of Amsterdam zu.

TESTE DEIN WISSEN

Independent Group 

Lösung anzeigen
TESTE DEIN WISSEN

separate groups of participants placed into different conditions of independent variable

1. Posttest-only design

​2.Pretest/posttest design

Lösung ausblenden
TESTE DEIN WISSEN

Bias responses: Socially Desirable Responding/ Faking Good (/bad)

Solution

Lösung anzeigen
TESTE DEIN WISSEN
  • Ensure anonymity
  • Include survey items that identify socially desirable responders, ask friends to rate, 
  • Computerized measures to evaluate opinions

 

Lösung ausblenden
TESTE DEIN WISSEN

Validity of frequency claims


Lösung anzeigen
TESTE DEIN WISSEN

Construct validity

Lösung ausblenden
TESTE DEIN WISSEN

Bias Responses 

Lösung anzeigen
TESTE DEIN WISSEN

= bad construct validity

1. Response Sets (= nondifferentiation) --> Shortcut taken when answering survey

People may adopt a consistent way of answering questions


  • Acquiescence=Yes, or strongly agreeing with items instead of critical thinking
    • Solution: Reverse-worded items (formulating it negative)
  • Fence sitting = Always in the middle of scale => hard to tell if true
    • Solution: Take away neutral options (may be invalid, when there is truly no opinion)


2. Socially Desirable Responding/ Faking Good (/bad)


3. Self-reporting more than they can know --> They can report how they feel/ think etc. but not why


4. Self-reporting memories: 

  • people's accounts for adverse events trusted
  • certainty, vividness doesn't mean accuracy
Lösung ausblenden
TESTE DEIN WISSEN

Correlational study

Lösung anzeigen
TESTE DEIN WISSEN

Positive association 

  • High scores on one variable go with high scores on the other variable 
  • & Low scores on one variable go with low scores on the other variable
  • --> Line in scatterplot incline upwards/ slope is positive

Negative association 

  • High scores on one variable go with low scores on the other variable 
  • & Low scores on one variable go with high scores on the other Variable
  • Line in scatterplot incline downwards/ slope is negative


Zero association 

  • no association between the two variables
  •  Line in scatterplot would be horizontal/ slope of zero

 

  • Stronger relationship = better prediction
  • Weaker relationship = some but uncertain prediction
  • Zero correlation = no predictions





Lösung ausblenden
TESTE DEIN WISSEN

External validity

Lösung anzeigen
TESTE DEIN WISSEN

How well the results of a study are generalizable to or represent people or contexts besides the original study.

  • Generalizability = How well did the researchers choose the study’s participants and how well do they represent the population intended to make claims about.
Lösung ausblenden
TESTE DEIN WISSEN

4 Different kinds of research

Lösung anzeigen
TESTE DEIN WISSEN
  1. Surveys / interview research
    1. Send out questionnaires, conduct interviews
  2. Observational research
    1. Observe behaviour that you are interested in
  3. Correlational research
    1. Study the association between two variables
  4. Experimental research
    1. Manipulate the cause and study the effect
Lösung ausblenden
TESTE DEIN WISSEN

Internal validity

Lösung anzeigen
TESTE DEIN WISSEN
  • In relationship of variable (A) & (B) = extend to which A rather than some other variable (C) is responsible for changes in B
  • Are there alternative explanations for the statistical effect?
Lösung ausblenden
TESTE DEIN WISSEN

Statistical validity (/Statistical conclusion validity)

Lösung anzeigen
TESTE DEIN WISSEN

How well the numbers support the claim. The extent to which a study’s statistical conclusions are replicable, accurate and reasonable

  • Margin of error = Statistical figure, based on sample size, that tries to include true value in the population (41%, margin of error: +/- 2.6% --> true % = 38.4‑43.6%)
  • how strong, big the effect is 
  • precision of the estimate
    • point estimate = value of study is an estimate of that value in some population
  • statistical validity improves with multiple estimates(replications)
Lösung ausblenden
TESTE DEIN WISSEN

Association claims - Validity 

Lösung anzeigen
TESTE DEIN WISSEN

Construct validity

Lösung ausblenden
TESTE DEIN WISSEN

Frequency claims

Lösung anzeigen
TESTE DEIN WISSEN

Describe a particular rate or degree of a single variable

  • Claim how frequent or common something is
  • Always focus on one variable
  • Variables always measured, not manipulated
Lösung ausblenden
TESTE DEIN WISSEN

Bias questions

Lösung anzeigen
TESTE DEIN WISSEN
  • Leading Questions = Wording leads to particular response, suggest particular viewpoint
  • Double-Barrelled Questions = ask 2 questions in 1 => people might be responding to only 1 part or both
    • Solution: Keep wording as simple as possible
  • Negative Worded Questions cause confusions, as it doesn't measure true beliefs
    • Solution: ask questions both ways (= study internal consistency)
  • Question order influences responses
    • Solution: Different versions of survey with a different question order 
    • => should get similar results with each order
Lösung ausblenden
  • 7622 Karteikarten
  • 238 Studierende
  • 0 Lernmaterialien

Beispielhafte Karteikarten für deinen Research methods Kurs an der University of Amsterdam - von Kommilitonen auf StudySmarter erstellt!

Q:

Independent Group 

A:

separate groups of participants placed into different conditions of independent variable

1. Posttest-only design

​2.Pretest/posttest design

Q:

Bias responses: Socially Desirable Responding/ Faking Good (/bad)

Solution

A:
  • Ensure anonymity
  • Include survey items that identify socially desirable responders, ask friends to rate, 
  • Computerized measures to evaluate opinions

 

Q:

Validity of frequency claims


A:

Construct validity

Q:

Bias Responses 

A:

= bad construct validity

1. Response Sets (= nondifferentiation) --> Shortcut taken when answering survey

People may adopt a consistent way of answering questions


  • Acquiescence=Yes, or strongly agreeing with items instead of critical thinking
    • Solution: Reverse-worded items (formulating it negative)
  • Fence sitting = Always in the middle of scale => hard to tell if true
    • Solution: Take away neutral options (may be invalid, when there is truly no opinion)


2. Socially Desirable Responding/ Faking Good (/bad)


3. Self-reporting more than they can know --> They can report how they feel/ think etc. but not why


4. Self-reporting memories: 

  • people's accounts for adverse events trusted
  • certainty, vividness doesn't mean accuracy
Q:

Correlational study

A:

Positive association 

  • High scores on one variable go with high scores on the other variable 
  • & Low scores on one variable go with low scores on the other variable
  • --> Line in scatterplot incline upwards/ slope is positive

Negative association 

  • High scores on one variable go with low scores on the other variable 
  • & Low scores on one variable go with high scores on the other Variable
  • Line in scatterplot incline downwards/ slope is negative


Zero association 

  • no association between the two variables
  •  Line in scatterplot would be horizontal/ slope of zero

 

  • Stronger relationship = better prediction
  • Weaker relationship = some but uncertain prediction
  • Zero correlation = no predictions





Mehr Karteikarten anzeigen
Q:

External validity

A:

How well the results of a study are generalizable to or represent people or contexts besides the original study.

  • Generalizability = How well did the researchers choose the study’s participants and how well do they represent the population intended to make claims about.
Q:

4 Different kinds of research

A:
  1. Surveys / interview research
    1. Send out questionnaires, conduct interviews
  2. Observational research
    1. Observe behaviour that you are interested in
  3. Correlational research
    1. Study the association between two variables
  4. Experimental research
    1. Manipulate the cause and study the effect
Q:

Internal validity

A:
  • In relationship of variable (A) & (B) = extend to which A rather than some other variable (C) is responsible for changes in B
  • Are there alternative explanations for the statistical effect?
Q:

Statistical validity (/Statistical conclusion validity)

A:

How well the numbers support the claim. The extent to which a study’s statistical conclusions are replicable, accurate and reasonable

  • Margin of error = Statistical figure, based on sample size, that tries to include true value in the population (41%, margin of error: +/- 2.6% --> true % = 38.4‑43.6%)
  • how strong, big the effect is 
  • precision of the estimate
    • point estimate = value of study is an estimate of that value in some population
  • statistical validity improves with multiple estimates(replications)
Q:

Association claims - Validity 

A:

Construct validity

Q:

Frequency claims

A:

Describe a particular rate or degree of a single variable

  • Claim how frequent or common something is
  • Always focus on one variable
  • Variables always measured, not manipulated
Q:

Bias questions

A:
  • Leading Questions = Wording leads to particular response, suggest particular viewpoint
  • Double-Barrelled Questions = ask 2 questions in 1 => people might be responding to only 1 part or both
    • Solution: Keep wording as simple as possible
  • Negative Worded Questions cause confusions, as it doesn't measure true beliefs
    • Solution: ask questions both ways (= study internal consistency)
  • Question order influences responses
    • Solution: Different versions of survey with a different question order 
    • => should get similar results with each order
Research methods

Erstelle und finde Lernmaterialien auf StudySmarter.

Greife kostenlos auf tausende geteilte Karteikarten, Zusammenfassungen, Altklausuren und mehr zu.

Jetzt loslegen

Die all-in-one Lernapp für Studierende

Greife auf Millionen geteilter Lernmaterialien der StudySmarter Community zu
Kostenlos anmelden Research methods
Erstelle Karteikarten und Zusammenfassungen mit den StudySmarter Tools
Kostenlos loslegen Research methods