class test 2 (week 5-8) Flashcards
1
Q
what are the uses for data in the SPSS?
A
> management
> analysis
> storage
2
Q
what does data management in the SPSS refer to?
A
- defining variables
- coding values
- entering and editing data
- creating new variables
- recording variables
- selecting cases
3
Q
what does data analysis statistics refer to in the SPSS?
A
- univariate stats
- bivariate stats
- multivariate stats
4
Q
what are the two screens on the data editor on the SPSS?
A
> data view: previous screen
> variable view: used to define variables
5
Q
what are the 10 characteristics used to define a variable in variable view?
A
Name Type Width Decimals Labels Values Missing Columns Align Measure
6
Q
what occurs in the data entry process?
A
- define your variables in variable view
- enter data, the values of the variables, in data view
7
Q
what is name as a characteristic to define a variable in variable view?
A
-each variable must have a unique name of not more than 8 characters and starting with a latter
8
Q
what are some things to remember when using names as a characteristic in variable view?
A
try to give meaningful names
9
Q
what are the two formats for type as a characteristic to define a variable in variable view?
A
internal and output formats
10
Q
what do internal formats as a type of variable characteristic refer to?
A
> numeric
> string (alphanumeric)
> date
11
Q
what do output formats as a type of variable characteristic refer to?
A
> comma
> dot
> scientific notation
> dollar
> custom currency
12
Q
what are numeric variables?
A
numeric measurements and codes
13
Q
what do string (alphanumeric) variable contain?
A
words or characters; strings can include numbers but, taken here as characters, mathematical operations cannot be applied to them
14
Q
what is the max size of a string variable?
A
255 characters
15
Q
what are labels for variables?
A
-descriptors for the variables
16
Q
what is the maximum amount of characters for labels of variables?
A
-maximum 255 characters
17
Q
when are labels used for variables?
A
output
18
Q
what are values of variables?
A
- value labels are descriptors of the categories of a variable
- coding
19
Q
what are the options for missing values and variables?
A
> up to 3 discrete missing values
> a range of missing values plus one discrete missing value
20
Q
what are columns of variables in data view?
A
-columns sets the amount of space reserved to display the contents of the variable
21
Q
what does align set in data view of variables?
A
whether the contents of the variable appear on the left, centre or right of the cell in data view
22
Q
are numeric variables right or left hand justified by default in data view?
A
right-hand
23
Q
are string variables right or left hand justified by default in data view?
A
left-hand
24
Q
what are the levels of measurement?
A
> nominal
> ordinal
> interval
> ratio
25
in SPSS where are interval and ratio levels of measurement designated?
designated together at scale
26
what is the default measurement level fro string variables?
nominal
27
what is the numeric measurement level fro string variables?
scale
28
wat is an interview transcript?
written record of an interview that has been transcribed from the verbal conversation
29
what are the benefits of transcribing?
> aids data familiarisation
> opportunity to reflect on interviewing style
> record emotional expression noted during the interview
30
what does confidentiality focus on dealing with?
participant identity
31
how can researchers practice confidentiality?
-interview transcripts use a pseudonym, change names of organisation, places and other people mentioned
32
what is re-identifiable data?
identifying info is removed from data and replaced with code so they can be re-identified
33
what is the iterative process in qual data analysis?
moving between data collection and analysis
34
what is inductive process in qual data analysis?
looking for patterns or themes in data rather than imposing ideas onto the data
35
what are some important parts of qual data analysis?
- make constant comparisons within and across interviews
- move from raw data to codes, subcategories, categories and finally themes
- draw well support conclusions based on data collected to answer research question
36
what are some types of methods of qual data analysis?
> content
> thematic
> discourse
37
what are some examples of approaches to data analysis for qual research designs?
> narrative analysis
> grounded theory methodology
> interpretive phenomenological analysis
38
what is thematic analysis?
-method for identifying, analysing and reporting patterns/themes within data
39
what is involved in the process of thematic analysis?
> familiarising oneself with data
> breaking the data into codes
> grouping similar codes together to form categories with associated subcategories
> define categories and associated subcategories
> group categories together form themes
40
what are the 4 steps of thematic analysis?
- data familarisation
- generating initial codes
- developing categories
- generating themes
41
what is involved in the data familiarisation step of thematic analysis?
undertake an initial quick read of each interview transcript and record initial response to data
42
what is involved in the generating initial codes step of thematic analysis?
code each interview transcript by writing a summary word or phrase that captures key idea foe each line or sentence within transcript
43
what is involved in the generating initial codes step of thematic analysis?
code each interview transcript by writing a summary word or phrase that captures key idea for each line or sentence within transcript
44
what is involved in the developing categories step of thematic analysis?
- re-read transcripts with initial codes looking for similarities/differences within single interview and across all interviews
- begin by grouping initial codes
- develop hierarchy system of categories and subcategories
45
what is involved in the developing themes step of thematic analysis?
focuses on examining relationships between each of the categories developed and how they might be grouped too fit under overall theme
46
what are some tips for initial coding?
- keep codes close to participant words/or use their words (in-vivo codes)
- use active language
47
what are some tips for coding sub-categories and categories?
- subcategories must relate to category under which they fall
- not all participants will have info that fits into all categories/subcategories
- participants might have differing views on same topic this can provide Moore complete understanding of sub-categories
48
what are some trustworthiness strategies?
- field notes
- audit trail
- triangulation
- member checking
- peer review
- prolonged engagement
- reflexivity
- thick description
49
what is the idea of probability?
everyone in pop will have the opportunity to have data collected
50
what are the three categories of non-respondents?
1. data collection procedures do not reach, not giving them chance to answer questions
2. asked to provide data who refuse to do so
3. unable to perform task or provide info needed
51
how is the response rate calculated?
number of people responding divided by number of people sampled
52
what is the response rate usually reported as?
% of a selected sample where data was collected or completion rat
53
will completion rates be lower or higher than response rates?
higher
54
how is non-response to surveys based?
non-respondents are systemically different from the whole pop
55
what is the response rates of mailed surveys?
5% to 20%
56
are response rates higher or lower in rural compared to city?
higher
57
what do acceptable response rates range from?
15-75%
58
what are some drawbacks of mailed surveys in regard to response rates?
- people more interested in topic likely to return= bias
| - better educated people often send back more quickly
59
what time of the day will telephone survey information be more distinctive
f data is collected between 9am and 5 pm on Sunday > Thursday
60
what are groups the tend to be under-represented in surveys?
```
> unemployed
> single
> recent migrant
> inner city areas
> low income
> low education
> don’t speak English
```
61
what are two issues that must be addressed to achieve high rates of response?
Gaining access to the selected individuals
| Enlisting their cooperation
62
how do you approach increasing response rate due to lack of availability for telephone surveys?
- make numerous calls during evening and weekend
- 6 to 10 calls per household often needed
- have interviewers with flexible schedules to make it convenient to respondents
63
how can you increase response rate by enlisting cooperation?
- send info letter in advance
- effectively present purpose
- ensure respondents are not threatened
- effective interviewers
64
what amount of people who initially refuse to survey involvement later agree?
approx .25 to .33
65
what can help increase response rate of mailed surveys?
- look more professional or personalised
- clear layout
- questions should be attractively spaced, easy to read and uncluttered
- tasks should be easy to do
- responses should be a check box, circle a number or other simple task
66
what is the process of sending out a mailed survey to increase response rate?
- about 10 days later mail reminder
- about 10 days after reminder send out reminder letter to emphasise importance
- last step if to contact respondents by telephone or email
67
what is a way to track who has or hasn't responded to mailed survey?
an identification number can be written on questionnaire or return envelope
-good practice to tell them what number is for
68
how can we enhance response rate of internet surveys?
make task easy, repeating contacts, using more than one mode to contact respondents and offering alternative modes of responding for this who don’t initially respond appear to maximise rates
69
what are three approaches to minimise resulting error when correcting nonresponse?
1. Using proxy respondents
2. Doing stat adjustments
3. Re-surveying sample of non-respondents
70
how are proxy respondents utilised to enhance response rate?
- many surveys collect data from one household respondent about other members
- if respondent is unable/unwilling to be interviewed asking another household member to report is an option
71
what type of information gathering is proxy respondents appropriate for?
factual but no subjective info
72
what are some things to take into account when planning your analysis?
budget, personnel, timeline, topic, target audience
73
what is the formatting technique for quant data?
> must quantify data
| > convert (data reduce) from collection format into numeric database
74
what is the formatting technique for qual data?
> must process data (type/enter-describe)
| > convert from audio/video to text
75
what is the formatting process of a quant and qual combination technique?
> process each element as appropriate
76
what are types of quant data analysis?
> counts, frequencies, tallies
| > stat analyses ( as appropriate)
77
what are types of qual data analysis?
> coding, categories
| > patterns, themes, theory building
78
what are types of qual and quant combination data analysis?
> process each element as appropriate
79
what are types of qual and quant combination data analysis?
> process each element as appropriate
80
what must occur before analysis?
must quantify data
81
what is quantification?
the process of converting data to numeric format
82
what are some simple transformations when quantifying?
> assign numeric rep to nominal or ordinal variables:
> turning male into 1 and female into 2
> assign numeric values to continuous variables:
> turning born in 1973 to 35
> number of children to 02
83
what is the goal of coding quant data?
reduce a wide variety of info to a more limited set of variable attributes
84
what is the purpose of codebook construction?
-guide data set construction, locating variables, and interpreting codes in data file during analysis
85
what does optimal scan sheets do when entering data?
limits possible responses
86
when is data entered in a CATI system/online
while collected
87
what is data entered directly onto in CATI system/online?
SPSS data matrix, excel, or ASCII file
88
what do CATI system/on-line typically work off?
coded questionnaire
89
what are the measures of central tendency?
mean, median, mode
90
what are the goals of presenting univariate data?
> provide reader with the fullest degree of detail regarding data
> present data in manageable form
> simple and straightforward
91
what is the point of subgroup comparisons?
describe subsets of cases, subjects or respondents
92
what is the bivariate analysis?
describe a case in terms of two variables simultaneously
93
how can you construct bivariate table?
- divide cases into groups according to the attributes of the independent variable
- describe each subgroup in terms of attributes if the dependent variable
- read table by comparing the IV subgroups in terms of a give attribute of the DV
94
does the dependent variable in a bivariate table go in the row or column?
row
95
does the independent variable in a bivariate table go in the row or column?
column
96
what type of variables are typically made into categorical variables in bivariate tables?
continuous
97
what are levels of measurement for continuous variables?
interval and ratio
98
what are four types of quant research methods?
- descriptive
- correlational
- cause-comparative
- experimental
99
what is involved in descriptive quant research method?
collecting data for hypothesis testing
100
what is involved in correlational quant research method?
determining whether and to what degree a relationship exists
101
what is involved in cause-comparative quant research method?
establishing the cause-effect relationship
102
what is involved in experimental quant research method?
establishes the cause-effect relationship, but manipulates the cause
103
wha are common methods of data collection in quant research?
> surveys/questionnaires
> structured interviews
> observation or interaction analysis
> secondary data or content analysis
> experiments
104
what is the purpose of descriptive stats?
to describe sample at hand
105
what do inferential stat procedures let us do?
generalise our findings beyond particular sample to larger pop
106
what are the three levels of descriptive stats?
> central tendency measures
> variability measures
> frequency and percentages
107
what are the four levels of measurement?
- nominal
- ordinal
- interval
- ratio (scale)
108
what is the nominal level of measurement?
basic clarification data;
| > don’t have meaningful numbers attached, but are broader categories
109
what is the ordinal level of measurement?
have numbers attached in certain order, not equal intervals between numbers
110
what is the interval level of measurement?
have equal intervals between numbers; distance between attributes have meaning
111
what is the ratio level of measurement?
have equal intervals between numbers absolute zero is meaningful
112
what are types of descriptive stats?
```
> number
> frequency count
> percentage
> deciles and quartiles
> measures of central tendency (mean, midpoint, mode)
> variability
> variance and standard deviation
> graphs
> normal curve
```
113
what is mode?
most frequently occurring value in a distribution (any scale, most unstable)
114
what level of data is mode best used for?
nominal
115
what is median?
midpoint in the distribution below which half of the cases reside
116
what level of measurement is median used for?
ordinal or above
117
is median sensitive to extremes?
insensitive
118
what is mean?
arithmetic average- the sum of all values in a distribution divided by the number of cases
119
what levels of data measurement is mean used for?
interval or ratio
120
is mean influenced by extremes?
yes
121
in a symmetrical distribution what falls at the same point?
median, mode, and mean
122
if a distribution is skewed to the right is it positive or negative?
positive
123
if a distribution is skewed to the left is it positive or negative?
negative
124
how are mode, median and mean related in a positively skewed distribution?
the mode is smaller than the median, which is smaller than the mean.
The median is the point x-axis that cuts the distribution in half, such that 50% of the area falls on each side
125
Does a hard test skew distribution to right or left?
left
126
Does a easy test skew distribution to right or left?
left
127
how is mean, mode, median related in a negatively skewed distribution?
mean being smaller than the median, which is smaller than the mode
128
what are some factors varying of results?
- dispersion
- range
- variance/standard deviation
129
what is dispersion?
extent of scatter around the average
130
what is range?
highest and lowest scores in a distribution
131
what is variance/standard deviation?
spread of scores, the greater the scatter the larger the variance
132
is standard deviation sensitive to extremes?
yes
133
what does standard deviation reflect?
how much subjects differ from the mean of the group
134
what does inferential stats allow for?
comparisons across variables
-hypothesis testing
135
what is the level of significance?
is predetermined level at which null hypothesis is not supported. The most common level is p < .05
136
what is an error type I?
reject the null hypothesis when it is really true
137
what is an error type II?
-fail to reject the null hypothesis when it is really false
138
what type of stats used to make decisions and report probability when we have made a type I error?
inferential stats
139
what does reporting the p value to readers alert?
the odds that we were incorrect when we decided to reject the null hypothesis
140
what levels of measurement are used in a chi-square test of independence?
two variables (nominal and nominal, nominal and ordinal, or ordinal and ordinal)
141
what are chi-square tests of independence affected by?
number of cells, number of cases
142
what type of hypothesis does a 2-tailed distribution chi-square test of independence result in?
null
143
what type of hypothesis does a 1-tailed distribution chi-square test of independence result in?
directional
144
what is correlation in inferential stats?
the extent to which two variables are related across a group of subjects
145
what is the range of a Pearson r relating to correlation in inferential stats?
-1.00 to 1.00
146
what does a 0.00 score indicate in a Pearson r relating to correlation in inferential stats?
absence of relationship
147
what does a t-test test for?
difference between two sample means for significance
148
what does analysis of variance (ANOVA) test for?
difference/s among two or more means
149
what type of analysis is used for ANVOA tests?
regression analysis (including step-wise regression)
150
what are the two types of stat presentations of data?
graphical and numerical
151
what is the graphical presentation of data?
ook for overall pattern and for striking deviations from that pattern. Over all pattern usually described by shape, centre, and spread of the data. An individual value that falls outside the overall pattern is an outlier
152
what are two types of data presentations used for categorical variables?
bar diagram and one charts
153
what are data presentation methods used for numerical data?
histogram, stem, leaf and box-plot
154
what is a bar graph?
lists the categories and presents the % or count of individuals who fall in each category
155
what is a pie graph?
lists the categories and presents the % or count of individuals who fall into each category
156
what is a histogram?
overall pattern can be described by its share, centre, and spread. The following age distribution is right skewed, the centre lies between 80 to 100, not outliers
157
what is a box plot?
describes the five number summary
158
what are the 5 approaches to instrument selection?
```
Standardised tests/performance measures
Contextualised/enviro assessment
Focus groups
Biometric measures
Document/record reviews
```
159
what do semantic differential scales ask?
respondents to rate given concept on a series of bipolar adjectives that are used to characterise one’s feelings, attitudes or reactions
eg. Good vs bad, dull vs exciting
160
what is the goal of retesting?
goal of protesting is to improve research instrument rather than to provide descriptions of pop
161
what can pretesting the sample design indicate?
whether design is feasible, provide an assessment of its difficulty and give a rough estimate of time and costs
162
what is sampling?
process where subgroup of participants is selected for study from larger group
163
what is random bias?
> happens by chance
164
what are some causes of systemic bias?
> volunteers
> non-respondents
> people that share common characteristics
> people with vested interest
165
what are some issues that can affect sampling?
- study questions/hypotheses
- research approach
> quant, qual, single-case design
-practical considerations
> availability, time, space, budget, etc.
-other issues:
> gender, age, location, SES etc.
166
what are some sampling techniques?
- volunteers/convenience sampling
- targeted recruitment
- inducement/compensation
- randomised: simple/stratified
- purposive
- snowball/networking
167
what are the sample requirements?
> representative
> specify inclusion and exclusion criteria
> establish min number of participants required
168
what occurs with probability sampling?
> each person can have equal chance of being selected
> pop known
> reduces sampling bias/increase study rigour
169
is the population known in probability sampling?
yes
170
when is non-probability sampling used?
> pop is unknown
| > not feasible to use probability sampling
171
what occurs in simple random sampling?
- sampling without replacement
- manual methods: draw out of hat
- random tables or computer generated lists
- use computer randomisation assignment programs
172
how is sample selected in stratified random sampling?
selection from identified subgroups
173
how is sample selected in systemic random sampling?
selection from every nth person on a list
174
how is sample determined in cluster sampling?
select groups or programs
175
what is purposive sampling?
-deliberate selection and recruitment
176
when is quota sampling used?
-when different proportions of participants are needed based on specific criteria
177
when is non-probability sampling used?
- clearly define process of sampling
- acknowledge limitation of sampling procedure used
- justify if the sampling limitations do harm the research question being answered
178
what are four important factors in sampling in quant research?
Sample size
Effective size
Level of sig.
Power
179
what is the p-value standard level used in most studies for the level of significance?
0.05
180
what is power calculation used for in sampling?
determines minimum sample size required
181
what is usually the power calculation used in OT studies
0.8
182
what are two principles of sampling in qual research?
-appropriateness and adequacy
183
what is appropriateness as a principle in qual sampling?
identification of participants who will best inform research about phenomena under inquiry
184
what is adequacy as a principle in qual sampling?
- enough data will be available to provide rich description of phenomena of interest
- continue to collect data until data saturation is reached or data source is exhausted (diminishing returns)
185
what are some strategies for sampling in qual research?
- purposeful selection
- max variation
- homogenous
- theory-based selection
- convenience selection
- snowball or network selection
186
what are common types of sampling in qual research?
- purposeful
- convenience
- theoretical
- max variation
- homogenous
- snowball
187
what is purposeful sampling in qual research?
-info rich participants
188
what is theoretical sampling in qual research?
- sampling used in grounded theory
| - seeks people, events, or info to refine theoretical categories emerging from data
189
what is max variation sampling in qual research?
- heterogenous (diverse) sample based on range of characteristics
- focus on exploring diversity and capturing breadth
190
what is homogenous sampling in qual research?
-selecting participants who are as similar as possible
191
what is snowball sampling in qual research?
-existing participants suggest new participants who meet criteria
192
when is homogenous sampling useful?
when resources are limited
193
what is generalisability?
To be able to accurately draw conclusions about the population by studying a subgroup or
sample of that population
194
what is effect size?
is the effect of difference between 2 means or the degree of correlation
between 2 variables in the results of a study
195
what is effect size?
is the effect of difference between 2 means or the degree of correlation
between 2 variables in the results of a study
196
what is referred to as the p value?
Probability statistic
197
what is power?
is the calculation completed to determine the minimum sample size required to
complete a study
198
what does reliability relate to?
how consistently items of a questionnaire measure what they claim
to measure
199
What type of questions is it when asked to self-reflect on experience, opinions, thoughts,
ideas, attitudes, or needs and then select the best option from a finite number of response
categories?
closed
200
what does a pilot study allow?
the examination of the procedures used when completing a survey;
it basically is a dry run of doing a larger survey, but instead mimicking it on a smaller scale.
