Research and Assessment Flashcards
(131 cards)
1
Q
What are the three types of research?
A
1) Qualitative Research
2) Quantitative Research
3) Mixed Methods Research
2
Q
What is qualitative research?
A
This type of research looks at characteristics of a population that are not number based, such as culture and social science issues. It can be a written as flexible report, talking about emerging themes.
It’s based on data, interviews, outreach, firsthand observations, and general themes (inductive).
3
Q
What is inductive reasoning?
A
This type of reasoning seeks to form a theory based on observations and findings.
It takes specific observations to create a general theory.
(E.g., All tests have been easy so far. Therefore, the next test will be easy.)
4
Q
What is deductive reasoning?
A
This form of reasoning seeks to test existing theories or hypothesis to prove it valid/invalid.
It takes a general theory and makes a specific inference from it.
(E.g., All girls like blue. –> Kim is a girl. —> So she likes blue). If this is proven false, the theory is wrong.
5
Q
What is quantitative research?
A
This type of research is number-based. It is written in a structured report.
Numerical data is analyzed to explore relationships between variables using statistical procedures (deductive).
6
Q
What is mixed methods research?
A
Combines qualitative and quantitative research for a more holistic analysis than using one method alone. Looks as both philosophical and theoretical frameworks.
7
Q
What is a Case Study Method? Is it qualitative or quantitative?
A
This research method looks at a specific instance/project/initiative for a detailed examination of it. It can be used as an illustrative example, but it is not used to compare two groups together.
Qualitative
8
Q
What is a Comparative Analysis? Is it qualitative or quantitative?
A
This research method compares different variables within the same setting/time frame to see how they are similar/different.
Qualitative
9
Q
What is a Discourse Analysis? Is it qualitative or quantitative?
A
This qualitative research method examines verbal, communicative, language, gestures, and written language, and any significant semiotic (signs/symbols) events. It studies how interpretations of the world/society/events/psyche are created based on language/communication.
10
Q
What are three forms of Discourse Analysis and what are they?
A
1) Semiotics - studies meaning of symbols
2) Deconstruction - studies relationship between text and meaning
3) Narrative Analysis - Studies stories/interviews/journals to understand chain of events/viewpoints
11
Q
What is e-Research? Is it qualitative or quantitative?
A
This form of research examines social interaction in the e-infrastructure. It uses of information technology to support existing and new forms of research.
Qualitative
12
Q
What is Ethnography? Is it qualitative or quantitative?
A
This qualitative research methods examines people in their natural setting (think Jane Jacobs). It’s based on firsthand observations of day-to-day activities, interviews, interactions, and exploration within the study area’s natural setting. It examines how things are normally occurring.
13
Q
What is Field Research? Is it qualitative or quantitative?
A
This research method requires the person to go into the study area and evaluate their surroundings on a typical day/existing conditions.
Qualitative
14
Q
What is Grounded Theory? Is it qualitative or quantitative?
A
This is an inductive reasoning which uses specific data collection and analysis together to create theories based on findings.
Qualitative
15
Q
What are three steps of Statistical Process?
A
(1) collect data (e.g., surveys)
(2) describe and summarize the distribution of the values in the data set
(3) interpret through inferential statistics and statistical modeling (i.e., draw general conclusions for the population on the basis of the sample.)
16
Q
What is NOIR?
A
NOIR ranks the level of measurement of data types:
Nominal
Ordinal
Interval
Ratio
17
Q
What is nominal data? Is it qualitative or quantitative?
A
This data type is categorical (think NAME). Lowest level of measurement.
- Exclusive categories
- Order doesn’t matter
- Qualitative, not numerical (they can be assigned #s but they are only labels)
- Can be categorical or dichotomous (only two values)
- Only central tendency analysis that makes sense is mode
- Can assess frequency and range
Qualitative
18
Q
What is ordinal data? Is it qualitative or quantitative?
A
This data type is categorical but has numeric ranking value (think ORDER).
- Difference between categories is unknown, just the order matters
- Zero is arbitrary
- Used to measure attitudes and perceptions
- Likert Scale
- Can be counted and ordered, but not measured
- Can use all nominal data analysis, as well as median, percentile, rank order of coefficients
(e.g., satisfaction level survey ranks 1 to 5, the difference between a level 4 and 5 is not exactly known and may be different to people, we just know that a 5 is more satisfied than 4.)
Qualitative
19
Q
What is interval data? Is it qualitative or quantitative?
A
This data type is categorical AND has equal distance between each value (think has a known interval aka distance).
- Difference between values are measurable and equal
- Zero is arbitrary (zero does not mean no value)
- Interval means “space in between”
- Can be negative distance
- All ordinal tests, as well as mean, standard deviation, add/subtract (can’t multiply and divide).
(e.g., temperature, test scores, time, elevation)
Quantitative
20
Q
What is ratio data? Is it qualitative or quantitative?
A
This data type is the highest measurement of data (think ratio ends in 0 so there is a zero)
- Zero is absolute, it means none (can multiply/divide)
- The differences are equal and measurable
- All statistical tests are possible
Quantitative
21
Q
What are qualitative variables?
A
This type of variable does not have a numeric value. (e.g., household income, level of a pollutant in a river)
22
Q
What is a quantitative variable?
A
This type of variable has a numeric value/meaning.
23
Q
What is a continuous variable?
A
This type of variable has endless number of values, positive or negative, and can become specific with decimal places.
24
Q
What is a discrete variable?
A
This type of variable has a finite number of values, are whole numbers (integers), and cannot be negative. (e.g., number of events, number of people)
25
What is a dichotomous variable?
This type of variable, also known as binary variable, only has two values. Often coded as 0 or 1. (e.g., yes/no, positive/negative, etc.)
26
What is a population?
The total of an entity/group.
27
What is a sample?
A subset of the population. This partial analysis is common in planning due to constraints on getting data on the whole population.
28
What is Descriptive Statistics?
This describes the characteristics of the distribution of a population or sample.
- Central Tendency: Mode, Mean, Median
- Range of Dispersion: Range, Standard Deviation
(e.g., "on average, AICP test takers in 2018 are 30 years old"). The context will make clear whether the statistic pertains to the population (all values known).
29
What is Inferential Statistics?
This form of stats uses sample data to draw a conclusion about the entire population.
We determine things about the population based on what is observed in the sample.
For example, we could take a sample of 25 test takers and use their average age to say something about the mean age of all the test takers.
30
What is distribution?
The overall trend/shape of the data. Can be shown as:
- An ordered table
- Histogram - grouped into bins aka the bars in a bar chart
- Density Plot - a smooth curve
31
What is symmetry in distribution?
When there is a similar range below or above the mean.
32
What is skewness in distribution?
When there are more higher or lower values and curve is not symmetrical. Skewed data is undesirable. (think skewed TAIL).
33
When is data considered Skewed Right?
- Skewed Right = Positive Skew = tail is skewed to the right (curve is shifted left) - more values are above the mean (in the tail)
Mean is higher than the median
34
When is data considered Skewed Left?
- Skewed Left = Negative Skew = trail is skewed to the left (curve is shifted right) - more values are below the mean (in the tail)
Mean is lower than the median
35
What is the range of data?
Highest to smallest values
36
What is kurtosis?
The flatness/peak of the distribution curve.
37
What does a tall curve mean?
Low level of variance (small range of values).
38
What does a flat curve mean?
High level of variance among values (wide range).
39
What is the Gaussian Distribution?
Aka the Normal Distribution, the bell curve is symmetric.
This distribution is symmetric and has the additional property that the spread around the mean can be related to the proportion of observations.
More specifically, 95% of the observations that follow a normal distribution are within two standard deviations from the mean.
40
What are the types of Central Tendency?
mean, median, and mode
41
What is a weighted mean?
When greater importance is put on specific values (multiplied).
For example, when computing a measure for the mean income among a number of counties, the value for each county could be multiplied by the number of people of the county, yielding a population-weighted mean.
42
What is the preferred Measure of Central Tendency to represent data?
Median, outliers can skew mean.
43
What do Variance and Standard Deviation measure?
They measure how much values are dispersed/spread out from the mean.
44
What is the difference between Variance and Standard Deviation?
They both measure distribution and are based on the squared difference of the mean, but Standard Deviation is the square root of the Variance to normalize the data (make it the same unit as the original variable).
45
What is the degree of freedom correction?
This is used when using a sample to estimate the Variance or Standard Deviation. Because a sample of a population has errors and the mean is not exactly known, one is subtracted from the number of instances (n-1) to offset the potential error.
The variance will be slightly larger than when calculated with n.
46
What is the Coefficient of Variation?
Basically expresses the amount of variation in a dataset as a %.
It's used to compare different datasets and how varied they are. The lower the better (the less dispersion/variation/risk).
= standard deviation / the mean.
47
What is a Z-Score?
It basically measures how far a value is "dispersed" from the mean in standard deviation units. There's a whole chart you use that goes with this.
Z-score = (value - mean) / standard deviation
(e.g., a z-score of more than 2 would mean the observation is more than two standard deviations away from the mean, or, it is an outlier in the sense just defined.)
48
What is a Hypothesis Test?
A test to reject the null hypothesis or statement about the population.
```
H0 = null hypothesis (original theory)
Ha = alternative hypothesis (what you seek to validate by rejected the H0)
```
The alterative hypothesis can be one-sided or two-sided.
49
What is a one-sided vs. two-sided hypothesis alterative?
Alternative Hypothesis types:
One-sided = can be either larger or smaller than H0 but not both
Two-sided = both directions are considered (larger and smaller)
50
What is a test statistic?
Is it the process to test the hypothesis test.
A test statistic describes how closely the distribution of your data matches the distribution predicted under the null hypothesis of the statistical test you are using.
51
What is a sampling error or sampling distribution?
The error from sampling only a portion of the population.
They are the difference between the real values of the population and the values derived by using samples from the population.
52
What is a systemic error?
It is a model misspecification, which occurs because our model (or assumptions) are wrong. It is unrelated to the sample.
53
What is the standard error?
It indicates how different the population mean is likely to be from a sample mean. It tells you how much the sample mean would vary if you were to repeat a study using new samples from within a single population.
Uses the same format as the Standard Deviation.
54
What is the P-Value? What are the other two names for it?
A.k.a Significance or Type I Error.
In null-hypothesis significance testing, the p-value is the probability of obtaining test results at least as extreme as the results actually observed, under the assumption that the null hypothesis is correct.
You want this to be small, either 5% or 1%.
55
What kind of decision is the rejection of a null hypothesis?
A **statistical decision**.
56
What is a Confidence Interval?
It is a range around the sample statistic that contains the population statistic with a given level of confidence, typically 95% or 99%.
57
How do you calculate the Inter-Quartile Range?
Q3-Q1 on a box whisker plot
58
On a box whisker plot, what is considered an outlier?
Values below Q1 or above Q3. This correlates with values below standard deviation of -2 or above 2.
59
What is a T-Test or Student T-Test?
This is a statistical test used to compare the [means](https://www.scribbr.com/statistics/mean/) of two samples. You use it to assess whether there is no difference between the two samples (accept H0) or they are different from one another (reject H0).
You compare the sample means. You want a confidence interval of 95% or higher.
* The null hypothesis (H0) is there is no difference between the groups (there is a significant relation/they are equal). Difference = 0.
* The alternate hypothesis (Ha) is that they are different (not related). Difference is not 0.
Ex: You want to know whether the mean petal length of iris flowers differs according to their species. You find two different species of irises growing in a garden and measure 25 petals of each species. You can test the difference between these two sample groups using a t-test.
60
When do you use a T Test?
A t-test can only be used when comparing the means of two sample groups to see if there is a significant difference.
A T Test is often used to **test the significance of a regression coefficient**
61
In a T-Test, when do you reject the hypothesis?
When t-value is greater than the critical value. You determine the critical value using the T Table with p value (p = 0.05) and the degree of freedom.
You reject the hypothesis if it is higher than the critical value → means the groups are different.
62
What are the 3 different kinds of T Tests and when do you use them?
**One-sample, two-sample, or paired t-test**
* If the groups come from a single population (e.g. measuring before and after an experimental treatment), perform a **paired t-test**.
* If the groups come from two different populations (e.g. two different species, or people from two separate cities), perform a **two-sample t-test** (a.k.a. **independent t-test**).
* If there is one group being compared against a standard value (e.g. comparing the acidity of a liquid to a neutral pH of 7), perform a **one-sample t-test**.
63
What is a regression coefficient?
Regression coefficients are values that are used in a regression equation to estimate the predictor variable and its response.
The sign of a regression coefficient tells you whether there is a positive or negative correlation between each independent variable and the dependent variable. A positive coefficient indicates that as the value of the independent variable increases, the mean of the dependent variable also tends to increase.
The null hypothesis is that the population regression coefficient is zero and the alternative coefficient is that it is non-zero. Rejecting the null hypothesis is interpreted as designating the coefficient as significant (at a given p-value). To compute the t-test in this case, we take the estimate and divide it by its standard error.
64
What is a critical value?
This is the value that marks to limit of the confidence area to accept the significance (validity) of the data or reject as random (invalid).
For a 95% confidence interval, you'd want the critical value to be based off of p = 0.5.
65
What is the difference between a one-tail vs. two-tail T-Test?
One-Tail Test = only examine whether higher OR lower (not both) - Are girls shorter than guys?
Two-Tail Test = examine the difference (could be either higher/lower) - Are girls' heights different than guys?
66
What's the difference between a paired or unpaired T-Test?
Paired = the values are from the same person/source at different times (e.g. two datasets of the test scores of each student before and after the lesson, the test scores are paired to the person)
Unpaired = the values are not related (E.g., datasets of test scores of girls and guys)
67
What is the Degree of Freedom?
Correction of sampling - It's the number of variables in each dataset minus the number of datasets. (e.g. df = N1 + N2 - 2)
Use this with the P value in the T Table to determine the critical value.
68
What is an ANOVA test?
It stands for Analysis of Variance test.
It's an extension of a T-Test (compares 2 datasets), where ANOVA compares 2 or more datasets.
It just shows if there is a difference, not where it occurs (see post-hoc test).
69
What is a post-hoc test?
This test is used after an ANOVA test to tell where the difference occurred.
Post-hoc is Latin for “after event”
70
What does a T-Test and ANOVA test assume?
That the data follows a normal distribution and the variances of the 2+ groups are equal.
71
What is a Chi-Square Test?
A “goodness of fit” test.
Tests whether a sample group matches a population.
A Chi Square test is often used to test the null hypothesis of independence in a contingency table, i.e. when the observations are grouped according to two categorical variables. The observed proportions are compared to the proportions we would expect if the two classifications were independent.
72
What is the Chi Square Distribution?
The **Chi Square distribution** is a skewed distribution that is obtained by taking the square of a standard normal variable (so, it only takes positive values). Under the null, the Chi Square test follows a Chi Square distribution.
73
What is a Bivariable Analysis?
Assess whether two variables are related to each other.
74
Correlation Coefficient
Measures the strength of which two varies are related to each other - only in a linear relationship. (Does NOT show causation i.e., whether one variable influences the other)
The correlation is computed by standardizing each of the variables and its value is between -1 and +1. The square of a correlation coefficient is often referred to as r2 (or R2), i.e., **r-squared**.
The correlation coefficient is typically used as a descriptive statistic, but it can also be construed as a **hypothesis test**. The null hypothesis of no correlation corresponds to a value of 0. A significant difference from 0 would suggest a linear relationship. A typical one-sided alternative hypothesis would be **positive correlation** (high values of one variable match high values of the other, and low values match low values), or **negative correlation** (high values of one variable match low values of the other, and vice versa). In practice, using the correlation as a hypothesis test is not that useful, since it almost always rejects the null hypothesis.
75
Correlation Coefficient
Measures the strength of which two varies are related to each other - only in a linear relationship. (Does NOT show causation i.e., whether one variable influences the other)
The correlation is computed by standardizing each of the variables and its value is between -1 and +1. The square of a correlation coefficient is often referred to as r2 (or R2), i.e., **r-squared**.
The correlation coefficient is typically used as a descriptive statistic, but it can also be construed as a **hypothesis test**. The null hypothesis of no correlation corresponds to a value of 0. A significant difference from 0 would suggest a linear relationship. A typical one-sided alternative hypothesis would be **positive correlation** (high values of one variable match high values of the other, and low values match low values), or **negative correlation** (high values of one variable match low values of the other, and vice versa). In practice, using the correlation as a hypothesis test is not that useful, since it almost always rejects the null hypothesis.
76
What is R-Squared?
The square of a correlation coefficient
77
What is linear regression?
This hypothesizes a linear relationship between a **dependent variable** (on the left-hand side of the equal sign) and one or more **explanatory variables** (on the right-hand side of the equal sign).
For example, a typical regression equation would be expressed as **y = a + b1x1 + b2x2 + e**. In this expression, y is the dependent variable, say the outcome on the AICP test, and x1 and x2 are explanatory variables, such as the number of hours studied and the years of experience. The e stands for a random error term, since the variables observed are a sample from the population. The coefficient is the **intercept**, and b1 and b2 are the **slope coefficients**. The coefficients of the linear regression are estimated by means of **least squares**, and their significance interpreted by means of a **t-test**.
78
What are the 7 types of population projection?
1. Linear Method
2. Exponential Method
3. Modified Exponential Method
4. Symptomatic Method
5. Distributed Housing Units Method
6. Step-Down Ratio Method
7. Cohort Survival Method
79
What is Linear Method?
Type of population projection where you take the average amount of growth and keep adding it (linear trend).
80
What is the Exponential Method?
You take the rate of growth and keep calculating it per year (it is a curve).
81
What is the Modified Exponential Method?
This type of population projection assumes growth eventually slows (looks like an S curve)
82
What is Gompertz Projection?
Building on the Modified Exponential Method, this population projection assumes growth is slowest in the beginning and speeds up over time (looks like an S too).
83
What is Symptomatic Method?
Projects population growth based on available data (like # of new housing units, # of permits, voter registration, utility connections, etc.)
For instance, with the average household size at 2.5, data on 100 new single-family building permits that are issued this year, would yield an estimate of 250 new people will be added to the community.
84
What is Step-Down Ratio Method?
Takes the proportion of a population of a larger area to calculate its future population growth.
This ratio is used to project the current or future population. For example, the population of Plannersville is 20% of the county population in 2000. If we know that the county population is 20,000 in 2005, we can then estimate the population of Plannersville as 4,000 (20%).
85
What is the Distributed Housing Unit Method?
This method uses the Census data for the # of housing units, which is then multiplied by the occupancy rate and persons per household.
Only good for stable communities, not rapid growing ones.
86
What is the Cohort Survival Method?
It uses the current population and natural increase (birth-death) and net migration (in migration - out migration) for men and women in specific age groups.
Provides the most accurate population projection but requires a large amount of data.
87
What is a population pyramid?
Shows the number of people per gender per age group (increments must be same)
Can be used for Cohort Survival Method:
Specific time intervals are used, such as one, five or ten years. The smallest time interval for which an estimate can be made is the length of time it takes for all members of an age cohort to age (e.g., age 10 - 14) to the next age grouping (e.g., age 15 - 19). All of the cohorts must have the same interval since each group must pass from one cohort to the next with nobody left behind over the course of the analysis. So, if data is available for each age (year), the method can be used to project the population year by year. Typically, five-year intervals are used. In that case, the shortest time for which a projection can be made is five years.
88
What is natural increase?
= Births - Deaths
89
Where are birth rates published?
by the state in the Vital Statistics of the United States through the U.S. National Center for Health Statistics.
90
What is the date rate?
Number of deaths per 1,000 people
91
What is the Crude Birth Rate?
Number of births per 1,000 people
92
What is the General Fertility Rate?
= Number of births per 1,000 females of child-bearing age
93
What is the Age-Specific Fertility Rate?
Number of babies born per 1,000 females in a certain age group
94
What is Net Migration?
= In-Migration - Out-Migration
95
What is Net Migration Rate?
= (In-Migration - Out-Migration) / total population
96
What is an Economic Base Analysis?
This analysis looks at basic or non-basic economic activities in the region. Examines location quotients.
The exporting industries make up the economic base of a region.
97
What is basic economic activity?
Export industries
98
What is non-basic economic activity?
Import industries
99
What is a location quotient?
Looks at concentration of an industry in a region compared to the nation and determines whether an industry is an import/export.
Location Quotient = industry's share in local region / its share of the nation (or other levels of government)
* Less than one = importing economy (smaller local industry)
* Greater than one = exporting economy (greater local industry)
* Equal to one = same as nation, neither import/export
100
What is a shift-share analysis?
Analyzes a local economy in comparison with a larger economy and changes in employment/GDP/etc. over a period of time.
* Looks at the differential shift, proportional shift, and economic growth.
* Determines what portions of regional economic growth or decline can be attributed to national, economic industry, and regional factors.
* Helps identify industries where the regional economy has competitive advantages over the larger economy.
101
What is an input-output analysis?
It examines the interdependence between various industries in an economy and their performance/output. Links suppliers/purchases to determine total output of a region.
An economy's total output is equal to total production plus intermediate sales.
* COSTLY - Requires a large amount of data
* Today done with computer software
102
What kind of suppliers are there in an input-output analysis?
Input-output analysis identifies primary suppliers, intermediate suppliers, intermediate purchasers, and final purchasers:
* Primary suppliers do not purchase input for production. They typically purchase only final goods;
* Intermediate suppliers sell outputs to either intermediate or final purchasers;
* Intermediate purchasers buy outputs from others and use them as inputs to produce outputs
* Final purchasers use their inputs as final goods.
103
What factors does the input-output analysis assume?
Input-output analysis assumes:
* no economies of scale
* that technology and labor are static
* that inputs are not substitutable
* that each industry only produces one group of goods
* Consumption of inputs is constant
* There are no national imports or exports
104
What does NAICS stand for? What is it?
**North American Industry Classification System (NAICS)**
Uniform classification system used by US/Canada/Mexico to compare different business statistics and industries in North America. Created by the Office of Management and Budget in 1997.
Has six digits that specify industry.
Used in shift-share analysis.
105
What is Deprication?
Reduction in value over time due to wear and tear
106
Acquisition Cost
The total cost of a individual or developer for purchasing a property.
In addition to cost of land, it incudes delivery charges, closing costs, fees, and other costs part of purchasing activity.
107
What is fair market value?
The price a property/asset would sell on the open market (what a buyer and seller would agree on)
108
What is a survey? Pros/cons?
Used to gage attitudes and characteristics of public on a range of topics. They take a sample of the survey.
109
What is a sampling frame?
The population of interest that is surveyed.
110
What is a cross-sectional survey?
This survey collects info in a specific point of time (e.g., POVs on downtown today)
111
What is a longitudinal survey?
This survey is conducted multiple times over a period (e.g., every year). It can be used to assess changes in attitude over time.
112
What is written survey? Pros/cons?
This survey is written, such as on a handout, newspaper, post card, etc.
* *Pros:**
- Good for surveying broad audiences, such as general opinions about a community.
- Cheap
* *Cons:**
- Takes long to get responses
- Low response rate (20%)
- Doesn't reach illiterate, non-English speakers, or senior populations
113
What is a group-administered survey?
This type of survey is for a small target group for a specific topic. You gather everyone together then survey them (e.g., end of a class).
## Footnote
* *Pros:**
- High and quick response rate
- Requires a small sample size
* *Cons:**
- May be hard to get everyone together
114
What is a drop-off survey? Pros/cons?
This survey is dropped off at residences or businesses, and they can respond on their own time.
Pros:
- Higher response rate than a mail-in survey because person to person contact
Cons:
- Takes a while to hear back
- Can be expensive to pay people to distribute door-to-door
- Sample surveyed is usually smaller than a mail-in survey
115
What is an oral survey? Pros/cons?
This survey is verbally given, over the phone or in person.
## Footnote
Pros:
- Can reach people unlikely to attend meetings
- Higher response rate than mail-in surveys
Cons:
- More expensive and labor-intensive than mail-in or internet surveys for people to call
- Interviews can be biased depending on the interviewer
- Long questions and multiple-choice q's are hard to administer
116
What are phone surveys? Pros/Cons?
This survey is conducted over the phone. Either as the main survey or a follow up to get more detail.
Pros:
- Useful for yes/no answers
Cons:
- More costly and labor-intensive than mail-in and online surveys
- Response rate greatly varies, depending on how many people you can reach
- Response rates are overall declining
- Interviews can be biased depending on the interviewer
- Long questions and multiple-choice q's are hard to administer
117
What is a sample design?
This is determining how the sample size is chosen and what the sample objective is.
Will it be **probability** or **non-probability** sampling?
118
What is probability sampling?
Where the sample size is representative of the population, there is a mathematical relation so precise conclusions can be drawn (with a confidence level).
Ex: 46% of the city's homeowners favor additional playgrounds for the city, with an error rate of +/- 2%.
119
What is non-probability sampling?
Where the survey is not necessarily representative of the population, e.g. sample size is too small. Conclusions must be taken with a grain a salt.
120
What are 4 types of probability (aka random) sampling?
1. Simple Random
2. Systematic
3. Stratified
4. Cluster
121
What is simple random sampling?
Everyone has an equal chance of being selected (independently selected)
Aka “**Method of chance Selection” or** “**Representative Sampling”**
122
What is systematic sampling?
A type of random sampling that targets a specific group, where every Xth person is selected. Population needs to be homogenous and uniformly distributed.
X = total population / sample size needed
123
What is stratified sampling?
A type of random sampling that targets a specific group, where the population is stratified into distinct categories where each has a sample taken out of it (ensures each category is represented).
This is more accurate than cluster sampling.
124
What is cluster sampling?
This type of random sampling targets specific groups, by taking only some clusters of a mix of the population to be your sample.
Pros: this saves times/cost compared to stratified sampling
Cons: it may not accurately represent the population, the mix in a cluster may not be the best representative
125
What is a convenience sample?
A non-probability type of sampling, where whoever is available at the time is surveyed.
126
What is Volunteered Geographic Information?
It is a non-probability type of sampling where people volunteer and give info attached to a location on a webmap (geospatial data) - think Waze/map.social.
127
What is volunteered sampling?
It is a non-probability type of sampling, where people self-volunteer to provide responses.
128
What is snowball sampling?
A non-probability type of sampling where the people you interview recommend new people to interview, and it keeps growing.
129
What is snowball sampling?
A non-probability type of sampling where the people you interview recommend new people to interview, and it keeps growing.
130
What is selection bias?
Means that the data (a sample from a population) selectively favors one fraction of the population, rather than being representative of the whole (bad)
131
What is implicit bias?
Implicit bias is a human trait, and is pervasive - refers to “the attitudes or stereotypes that affect our understanding, actions, and decisions in an unconscious manner."
The implicit biases we hold may run counter to our declared beliefs.
We do not even realize it.
