Brainscape's Knowledge GenomeTM

Browse over 1 million classes created by top students, professors, publishers, and experts.


See full index

1010 > PS1010 - Understanding Psychological Research > Flashcards

PS1010 - Understanding Psychological Research Flashcards

1
Q

If you aim to look for a difference between groups of participants or conditions in your study what research design is needed.

A

Experimental designs

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
2
Q

If you aim to look for a relationship between two continuous variables in your study what research design is needed.

A

Correlational designs

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
3
Q

If you aim to identify categories and themes from text based data in your study what research design is needed.

A

Qualitative designs

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
4
Q

What is a theory?

A

A broad & Developed idea that can explain something about human behaviour.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
5
Q

What is a hypothesis?

A

A specific prediction about an aspect of human behaviour that you can test in your study.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
6
Q

How do we research?

A
  1. Devise a research question
  2. Design a study to test the question
  3. Choose the stimuli, questionnaires, etc
  4. Apply for ethical approval
  5. Run the study and collect data
  6. Analyse data, see what you found out
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
7
Q

Why should we think about statistics when designing a piece of research?

A

Because research design and statistical analysis are strongly linked.
• Different research designs require different methods of analysis.
• Different types of data require different methods of analysis.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
8
Q

Rank some types of research in order of typically more qualitative techniques to typically more quantitive techniques.

A

Observational techniques
Interview and survey techniques
Correlational research
Experimental research

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
9
Q

What could an Experimental design involve?

A

Involves manipulating variables and looking for differences.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
10
Q

What is the Dependent and Independent variable?

A

The Independent Variable is the manipulation used by the researcher.
The Dependent Variable is the outcome measured by the researcher.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
11
Q

What is the difference between ‘Experimental design’ and a ‘Quasi experimental design’?

A

In an experimental design, participants can be randomly allocated to the different conditions.
However in a quasi experimental design, participants cannot be randomly allocated; groups are pre-defined.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
12
Q

What is an Independent measures design?

A

An Independent measures design has different groups of participants complete the experiment and are compared.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
13
Q

What is a repeated measures design?

A

A Repeated measures design is where participants repeat the experiment under different conditions.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
14
Q

Describe the variables in correlational designs.

A
  1. There is no ‘independent variable’ and no ‘dependent variable’. No variables are manipulated, naturally occurring variables are measured.
  2. Variables must be continuous scores (not groups).
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
15
Q

What is a continuous variable and give some examples

A

Score exists on a continuum:
• IQ
• Reaction times
• Accuracy

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
16
Q

Give some examples of categorical variables - both (2 groups: binary) and (3+ categories)

A

Categorical variables (2 groups: binary)
•Diagnosis: clinical or non-clinical group
•Verdict: guilty or innocent
•Handedness: left or right

Categorical variables (3+ categories): No linear order – place in any order!
•Drug type: placebo, aspirin, codeine
•Hair colour: blonde, brunette, redhead
•Degree: psychology, history, English

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
17
Q

What is important to remember for categorical variables with 3+ categories?

A

There is no linear order and so can be placed in any order.
E.g. Degree: Psychology, History and English
There is no order in which you must place these.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
18
Q

What are the four assumptions (both before and after data collection) to be met for a parametric design?

A

Assumptions met by good design (before data collection):

  1. Interval or ratio level data
  2. Independence of observations: No data point should influence another; all data should be independent.

Assumptions met by good data (after data collection):

  1. Data are (roughly) normally distributed
  2. Homogeneity of variance
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
19
Q

What types of non-parametric variables can we measure?

A

Nominal (categorical) data and Ordinal data

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
20
Q

What types of parametric variables can we measure?

A

Interval data and Ratio data

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
21
Q

What is Nominal (categorical) data?

A
  • Frequency of belonging to a ‘category’

* Example: Handedness, classed as left, right or mixed handed?

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
22
Q

What is Ordinal data?

A
  • Clear order to data, but distance between points may vary

* Example: Place in a race, 1st, 2nd, 3rd place etc…

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
23
Q

What is Interval data and give an example?

A
  • Order to data points, fixed distance between points and negative values
  • Example: Temperature, 1° is always the same, negative temperatures are possible
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
24
Q

What is ratio data and give an example?

A
  • Order to data points, fixed distance between points but no negative values
  • Example: Height, 1cm is always the same at all heights, no negative heights
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
25
Q

What is a sample?

A

All of the participants that we collect data from in our study

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
26
Q

What is a population?

A

All of the possible people that could have been included in our study

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
27
Q

What is a Null hypothesis (H0)?

A

Prediction that no effects will be found

28
Q

What is an alternative hypothesis (H1)?

A

Prediction that effects will be found

29
Q

What is a one-tailed hypothesis?

A

The direction of predicted effects is specified

30
Q

What is a two-tailed hypothesis?

A

Prediction that effects will be found, but with no direction predicted

31
Q

What is a Type I error?

A

When you find an effect in the result of your statistical analysis (the sample) but there is no effect or truth in the real world (the whole population).

32
Q

What is a Type II error?

A

When you don’t find an effect in the result of your statistical analyses but there is a finding/effect in the real world population that you missed.

33
Q

Which error is worse?

A

A Type II error happens when we have a study and find nothing in that data. However, if you were to look at the population, there is a finding that you missed. This error isn’t as bad because you can think about the sample you collected, the methodology, the confounds and what you can do to make the study stronger.

A Type I error is a really big problem in research, it’s when you find something from results of a sample, but it isn’t true for the population, so it’s a false positive.

34
Q

What is the alpha level?

A

The percentage risk of a Type I error

35
Q

What percentage of Type I error is acceptable?

A

5%

36
Q

What is a p value?

A

The probability of making a Type I error

37
Q

What is a significant result in terms of p values?

A

If the p value is less than 5% then we can say that it’s a significant result. It’s a big enough effect that we can be 95% confident that the findings are true and exist in the real world.

38
Q

What p values are significant and not significant?

A

Greater than .050 (p>.050) are not significant.

Less than or equal to .050 (p ≤ .050) is significant.

39
Q

What is a statistically significant result?

A

A finding within your sample of participants that you are at least 95% certain exists in the population too.

Less than 5% chance of you having messed up and saying that you found something that does not really exist!

40
Q

What direction is the percentage of Type I error in a one tailed hypothesis?

A

As you specify the direction, the entire 5% error is on that side of the data.

(If confused look at lecture 3 page 4 normal distributions)

41
Q

What direction is the percentage of Type I error in a one tailed hypothesis?

A

If you have a two tailed hypothesis, it’s open ended and you don’t have a directional prediction. You share the 5% error on either side.

42
Q

What are degrees of freedom (df) based on?

A

The number of participants (N-1)

The number of conditions (k-1)

43
Q

Why are more degrees of freedom better?

A

The more participants you have, the more likely it is that your sample reflects the population. You use degrees of freedom to tell you whether the calculated value is significant related to the number of participants you have. The more degrees of freedom there are have, the more participants there are and the more confident you can be that the findings are real. – less chance of a Type I error.

44
Q

What is the calculated value?

A

The value you calculate, how “big” the effect is. Bigger calculated values mean larger effects.

45
Q

What is a critical value?

A

The smallest value for which you can say the effect is significant.

46
Q

When is the calculated value significant?

A

If calculated value is larger than the critical value it’s significant.

47
Q

What four pieces of information can tell you the critical value and if your result is significant?

A
  1. Is your hypothesis 1 or 2 tailed?
  2. What is your alpha value?
  3. What is your calculated t value?
  4. What are your degrees of freedom?
48
Q

Why is the critical value different for increasing degrees of freedom?

A

Because a finding with a larger number of participants is more trustworthy so the critical value is smaller. Because there’s less chance of it being a mistake.

49
Q

Why is the critical value different when the alpha level gets smaller?

A

As the alpha level gets smaller the critical value is getting bigger because to be more confident you need a bigger effect.

E.g. at an alpha level of 0.01 you would be 99% confident which is why the critical value is bigger – you need a bigger effect to be more confident.

50
Q

Why is the critical value different for one and two tailed hypotheses?

A

1 tailed has a specific directional prediction and you are rewarded for this with a smaller critical value.

51
Q

What are the two stages to interpretation?

A
  1. Is the finding significant or not significant (NS)?
  2. If significant, what is the direction?
    (Which group has higher scores/ Is it a positive or negative relationship)
    But if the finding is not significant you don’t interpret the effects with this second question.
52
Q

What is wrong with the term insignificant?

A

It must be written as not significant (NS)

53
Q

What is a confounding variable?

A

This is a variable that we don’t manipulate or measure, but one that is still likely to somehow influence our findings.

For example, if you are researching whether lack of exercise leads to weight gain, then lack of exercise is your independent variable and weight gain is your dependent variable. Confounding variables are any other variable that also has an effect on your dependent variable.

54
Q

What other terms might be used to describe a confounding variable?

A

Control variable, covariate or mediating variable.

55
Q

What is an independent measures design?

A

When different participants take part in each condition in an experiment.

56
Q

What other term might be used to describe an independent measures design?

A

Unrelated

57
Q

What type of design is opposite, or should be compared to, an independent measures design?

A

Repeated measures design

58
Q

What is the difference between a frequency design and an experimental design - use examples.

A

To distinguish between the two, think about the data you collect. If it’s the number of people in a group then it’s a frequency design. However if it’s a continuous score such as IQ, reaction times etc, then it’s an experimental design.

59
Q

What do descriptive statistics do?

A
  • Easily summarise data
  • Reduces a large amount of numbers to a single representative number
  • Aids representation
60
Q

What types of variance are there?

A

Experimental variance and random variance.

61
Q

What is experimental variance?

A
  • Due to experimental manipulation
  • Variability between conditions
  • The variance that the researcher created through experimental manipulation and is measuring
62
Q

What is random variance and give an example?

A

• Due to measurement error
• Due to individual differences or unmeasured variables
(confounding variables)
• Variability within conditions

E.g. Measurement error could be the way the researcher is taking in results such as problems with a questionnaire that they created.
Individual differences might be the participants BMI, personality traits.

63
Q

What type of variance makes experiments more likely to be significant?

A

The more experimental variance there is, the more likely the results are significant.

Whereas the more random variance, the less likely it is for the results to be significant.

64
Q

What part of a report would you suggest why there’s variability in the experiment?

A

Use the discussion part of the report to explain why there is so much variability in an experiment. E.g. is there a measurement error or a confounding variable that could explain it.

65
Q

How do you present descriptive statistics?

A
  • Present descriptive statistics using either a table or a graph
  • Include a table/figure legend (Figure 1 with a brief description)
  • Refer to the table/figure within the text (e.g. see table one)
  • Integrate writing about descriptive with the writing about inferential statistics

1010 (5 decks)

Brainscape helps you reach your goals faster, through stronger study habits.
© 2024 Bold Learning Solutions. Terms and Conditions