PSYC2009 Flashcards

(95 cards)

1
Q

Organised scepticism

A

all evidence supports something but does not completely prove something e.g. homosexuality used to be proven as a disease and has now changed (not personally attached

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

why do people research?

A

EDEP
Exploratory purposes - discover and explore something
Descriptive - describe phenomena on how things are viewed
Explanatory - explain phenomena on how things are explained
Predictive - allows us to make predictions on what happens in the future

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

pilot test

A

get people from a small group of people you are going to survey and see if the data is similar to what you are expecting

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

Objective

A
  • A true answer exists
  • a participant can theoretically answer the questions accurately
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
5
Q

subjective

A
  • asking about personal perceptions
  • no actual true/false answer
  • cannot answer questions accurately
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
6
Q

Open-ended (unstructured) surveys

A
  • useful for gathering rich information
  • good for descriptive and explanatory work
  • more difficult and subjective to answer and therefore more time consuming
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
7
Q

Close-ended (structured) surveys

A
  • Pre-set responses
  • good for hypothesis testing
  • easy and quick
  • objective
  • may loose important information due to lack of options
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
8
Q

Dichotomous - Response format

A
  • two response options
  • e.g. yes or no
  • simplest type of quantification
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
9
Q

Multichotomous - Response format

A
  • choose all that apply
  • e.g. which pets do you have? dog, cat, otter
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
10
Q

Ranking - Response format

A
  • Ranking importance of several different options
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
11
Q

Verbal Frequency - Response format

A

subjective sense of how many times something has happened to them

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

Likert Scale - Response format

A
  • Agreement to non agreement
  • 2 to 11
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
12
Q

Semantic differential - Response format

A
  • Two words at opposite ends with interval marks
  • e.g. introvert at one side and extrovert at the other side
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
13
Q

Graphical - Response format

A

Mark your response on the line between two words

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

Non-verbal - Response format

A

shows your opinion without using words
e.g. point to a face that reflects your experience

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
15
Q

Things to avoid when writing a survey

A
  1. having to retain too much information
  2. have to perform mental calculations
  3. recall things in the past
  4. complex language
  5. requiring the person to gain new skills on the spot
  6. asking questions they have no prior knowledge too
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
16
Q

double barrelled questions

A

two concepts in one question

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

double-negative questions

A

never having to not

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

Leading questions

A
  • questions that suggest the answer the researcher is looking for
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
19
Q

Loaded questions

A
  • Questions that suggest socially desirable answers, or are emotionally charged
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
20
Q

Measurement error

A

the statistical deviation from the true value caused by the measurement procedure

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

observed score

A

true +/- measurement error

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

Two things error can be

A

systematic or random

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

no error in a line

A

straight line

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
24
consistent error
mostly straight line
25
non consistent error
bumps in the line
26
How to reduce measurement error
- do it online - minimise potential demand characteristics
27
Reliability
the consistency of: - items within the measure - the measure over time
28
Validity
whether the measure actually measures what it is intended to measure
29
Steps of Sampling
Target audience - who do you want to generalise in order to save time and cost Sampling frame - who has a chance of being selected Sample - who was selected Representativeness - to what extent is the sample a good indicator of the target population
30
Response rates influenced by
- rewards for people giving biased answers - ease of returning survey - follow up calls and presentation
31
Sampling Process
ISCM 1. Identify - Identify the target population and sampling frame 2. select - select a sampling method 3. calculate - calculate your sample size 4. Maximise - Maximise the return rate
32
Types of research
1. Experimental 2. Quasi-experiment 3. Non-experiment
33
Experimental
- active intervention - Random assignment - causal inference
34
Quasi-experimental
No Random Assignment
35
Non-experimental
- no manipulation - no casual inference
36
Levels of measurement
- Nominal/categorical - Ordinal - Interval - Ratio
37
Nominal/categorical
numerical labels (attributes are named)
38
Ordinal
Ordered numerical labels, with unclear intervals (distance is not meaningful)
39
interval
ordered with equal intervals in between (distance is meaningful)
40
ratio
continuous, with a meaningful 0 (absolute zero)
41
Descriptive stat vs inferential stat
Descriptive - describing the data Inferential - making inferences about the population based on the sample
42
Parametric stats
compare means (usually) - estimate the parameters of the population based on normal distribution - e.g. mean, STD, Skewness, Kurtosis
43
Non-parametric stats
Do not assume sampling from a population which is normally distributed - median, frequencies
44
Central Tendency
frequency, mode, median, mean
45
Distribution
spread: min, max, range, percentiles, variance, STD Shape: Skewness, Kurtosis
46
Distribution
Measures the shape, spread and dispersion of your data, deviation, central tendency
47
Variance
Average squared distance from the mean
48
Variance square root
Standard deviation
49
Standard deviation squared
variance
50
(X-Xbar)squared
sum of squares
51
x bar
sample mean
52
VAR P
Population
53
VAR S
Sample
54
Positive skew
start high and go low
55
Negative Skew
start low and go high
56
normal distribution
graph
57
Kurtosis
How flat vs how peaked the distribution of the data is
58
variance formula
variance formula
59
Positive kurtosis
Peaked data
60
Negative kurtosis
flat data
61
Examples of kurtosis
graph
62
Analysing non-normal distributions
1. How many peaks? 2. Is there a tail? 3. How peaked or flat is it (kurtosis)
63
Types of tails
positive skew - to the right Negative skew - to the left
63
Types of peaks
unimodal - one peak bimodal - two peak Multi-modal - 3+ peaks
64
Bounded scales
sometimes the highest or lowest scales are not considered
65
Non-parametric (graphs)
- bar graphs - pie charts
66
Parametric (graphs)
- Histogram - Stem and lead plot - Box plot
67
Lie factor (graph integrity)
Size of effect in graph divided by size of effect in data
68
Misleading graphs
- use perspective in misleading ways - leave out important context
69
Significance Testing
determine whether something significantly deviates from something that is chance
70
Null hypothesis (HO)
we always assume that the null is true and that it is not being affected by any other variable
71
Alternative hyptothesis (HA)
there is something else influencing the results (relationship between variables)
72
graph showing null hypothesis potentially true (small effect size)
Top left graph
73
Graph showing support to alternative hypothesis and big effect size
Bottom right graph
74
how small is the p-value
when the p-value is smaller than 0.05 its considered small
75
small samples lead to bigger or smaller p-value
larger p-value
76
two types of relationships
1. correlation 2. Regression
77
Effect size
the correlation coefficient or regression coefficient for them
78
P-value below 0.05
statistically significant in difference (reject Null)
79
P-value above 0.05
not statistically significant in difference (accept null)
80
Difference between groups can be also found through
1. t-tests 2. ANOVA
81
p-value depends on
- effect size - sample size - critical alpha
82
Confidence intervals
A range of values that contains a specified percentage of the sampling distribution
83
how to write confidence interval
mean = 53.0, 95% CI (48.2, 57.8)
84
equation for sample mean
population (u) + error (e)
85
t-statistic
The size of the difference between your sample mean and the population mean relative to the variation in the sample data
86
types of distribution
z,t,n distribution graph
87
effect of sample size on t and normal distribution
the sample size increases the difference between t and normal distribution decreases
88
halfwidth def
the distance between lower estimate to mean or higher estimate to mean
89
calculating Confidence Intervals
1. collect a sample of N observations 2. calculate the sample mean, STD, and standard error 3. calculate alpha 4. look up the two-tailed t-value that corresponds to N-1 5. calculating the half-width of the confidence interval (formula) 6. confidence interval statement
90
standard error of the mean
standard error of the mean formula
91
Sx is
standard error
92
S
sample standard deviation
93