click below

click below

Normal Size Small Size show me how

# Plan Making

### Visioning and Research Methods/Data Collection

Question | Answer |
---|---|

Visioning | Visioning seeks to answer 4 questions: 1) What is the current status of the community? 2) What is the current direction of the community? 3) What is the desired direction and future of the community? 4) How can this future be achieved? |

Visioning (2) | Visioning should attempt to incorporate community values; identify trends and pertinent issues; address needs from all areas of the community; encourage and facilitate local action; and set forth a desired future shape by the entire community. |

Visioning (3) | The most important aspect of visioning is public participation - a community is more likely to support a vision and plan that it helped create. |

Research - Quantitative | Quantitative - Statistical linear methods of analysis (linear regression and telephone interviews = ways of gathering quantitative data). |

Research - Qualitative | Opinions, emotions, themes, agendas, needs and interests. Qualitative data is normally garnered through interviews, workshops, and similar types of meetings, and is helpful for spotting trends and patterns. |

Validity | Internal validity - measures the reliability of a study, and indicates the extent to which a study considers all relevant variables. External validity - indicates the study's relevance to other situations and studies. |

Population | The entire group to which the data pertains, it's the focus of the research. A population normally shares some common characteristic, known as a "population parameter", which identifies and separates it from other populations. |

Population parameters | Race, socioeconomic status, place or residence, etc. |

Sample | A subset of the total pop and, ideally, is selected at random. Samples are taken when testing the entire pop is cost or resource prohibitive. Researchers gather statistical data on the sample, then use the data to project info of the entire pop. |

Convenience Sampling | A technique in which a researcher gathers data on only those test subjects that can be reached easily and readily. May produced biased results because the subject are not selected at random or without bias, and some subjects are excluded automatically. |

Snowball Sampling | A technique in which the researcher interviews an ever-increasing number of test subjects as the study progresses. The interviewer selects new interview topics and subjects based on the data gathered from previous subjects. |

Inferential Statistical Method | Inferential statistical method derives information about an entire population based on statistical data gathered from a sample of the population. |

Confidence Interval | A certain range of values around a sample statistic. When the estimated value (derived from sample statistic) is applied to the entire population, the actual value (within the population) should fall somewhere within this range. |

Confidence Interval (cont.) | For example, if the sample statistic is 50%, the actual value within the population may fall between 45% and 55%. This range is the confidence interval. |

Error Level | aka Margin of Error. The maximum possible difference between the estimated value gathered from the sample statistic and the actual value w/in the pop. For ex, a study may have a margin of error of plus or minus 3% from the actual real value w/in a pop. |

Confidence Level | Confidence Level measures the expected number of times the same set of results would be repeated if the same study was conducted 100 times. It is written as a percentage. |

Variables | A characteristic that is being measured w/in test subjects. Common variables include income, age, gender, family size, height, length, width, construction type, building decay, and monthly/annual costs. Three types of variables: Nominal Ordinal Interval. |

Nominal Variable | aka "categorical variables" are arranged into various population groupings, such as right-handed, left-handed, Republican, Libertarian, Baby Boomers, etc. |

Ordinal Variable | aka "ranking variables" are arranged within a certain hierarchical order. For ex, at the end of the season, sports teams are often ranked according to their win and loss records. A team with a 13-3 record is ranked higher than a team with a 5-11 record. |

Interval Variable | Are arranged along a numbered scale. The value difference btw intervals along a scale is numerical and progressive. Ex, gens can be measured according to their electrical output. One gen may put out 1,000 watts while a 2nd gen may put out 2,000 watts. |

Classification of Variables | Dependent, Independent, Treatment, Control, Confounding, Discrete, Continuous, Dichotomous |

Dependent Variable | Is manipulated by the researcher in order to affect an outcome that can be measured. In graphs, the dependent variable is normally along the y-axis |

Independent Variable | Independent variable is used to define or affect the behavior of a dependent variable. In graphs, it is normally placed along the x-axis. |

Treatment Variable | The treatment variable describes an independent variable that is being manipulated within the study. |

Control Variable | The control variable describes an independent variable that is not being manipulated within the study. |

Confounding Variable | A confounding variable is not one of the independent variables, but may be affecting the dependent variable. Conf variables offer an alternate explanation for the behavior of the dependent variable. |

Discrete Variable | A discrete variable has a finite number of values, and each value is an integer. An example includes the number of crayons in a box. |

Continuous Variable | A continuous variable has an infinite number of values. An example includes the speed of a moving object when expressed using fractions. |

Dichotomous Variable | aka "binary variable" is limited to 2 possible values, such as "on" or "off". |

Statistics | |

Mean | Average of all values related to a particular variable. Statistical outliers (extremely high or low measurements) often skew the mean. |

Median | Middle value within a distribution of measurements, and is less influenced by statistical outliers than the mean. Exactly half of the measurements will fall above the median, and exactly half will fall below. |

Mode | Measurement that occurs most often in a distribution. A bimodal distribution is one in which 2 values are tied for most occurrences. |

Range | The range is calculated by subtracting the lowest observed measurement from the highest observed measurements. |

Variance | The variance is the distribution (spread out or tightly clustered) of measurements. |

Standard Deviation | The standard deviation indicates the level of spread within a distribution. It is calculated by taking the square root of the variance. |

Distribution | The distribution is the entire set of a particular variable that has been tested for every subject in a population. For instance, if a study measured the height of 10 people, the distribution would consist of the 10 separate height measurements. |

Outlier | The outlier is a variable that falls well outside the grouping of most other variables within the distribution. For instance, if 9 people measured btw 5'9" and 6' in height and only 1 person measured 6'6" in height, that person would be the outlier. |

Frequency Distribution | Frequency distribution is the frequency with which measurements reoccur in a distribution. For instance, if 4 people measured in at 5'10" in height, the measurement of 5'10" would have a frequency distribution of 4. |

Chi Square | The chi square determines whether or not two variables are related. |

Linear Regression | Linear regression is used when a specific set of assumptions is met. It assesses the extent to which an independent variable affects a dependent variable. |

Scatter Plot Diagram | This can be used to show the relationship btw a dependent var. and an indepnt var. w/in a sample. Each measrmt taken from a test subject is shown as a point w/in the diagram. The points are connected to reveal patterns, which reveal tendencies w/in data. |

Scatter Plot Diagram Distribution Patterns | U-shaped pattern, J-shaped pattern, Bimodal pattern, Skewed pattern |

U-shaped pattern (scatter plot) | measurements are highly polarized |

J-shaped pattern (scatter plot) | measurements are concentrated on one end |

Bimodal pattern (scatter plot) | measurements are split between 2 primary values |

Skewed pattern (scatter plot) | measurements are not distributed equally |

Bell Curve | aka "normal distribution" indicates that the observed measurements occur symmetrical around the mean. 68% of measurements are located inside 1 standard deviation from the mean. 95% located inside 2 standard deviation from mean. 99% located 3 standard dev |

Statistical Analysis Tools | |

Population projection | Examines the current pop conditions in order to forecast future pop conditions. Only accurate if pop growth is not affected by some unforeseen external influence. |

Cohort Survival Method | Population projection method which divides the population into different age groups and then forecasts the distribution between these age groups over the next 5 or 10 years. |

Estimation Methods | |

Migration + natural increase method | Estimates population based on birth and death statistics. |

Ratio method | aka "step-down" method estimates the growth rate of a small area (i.e. neighborhood) using the growth rate estimation for a large area (i.e. City) |

Symptomatic method | Estimates current population based on easily accessible data such as school enrollment, building permits, and voter registration. |

Types of Areas | |

Census Tract | A small subdivision of a county. It is permanent, and used for gathering statistical data. Pop is around 4,000 people and consistent re: living, economic and pop conditions. Boundaries defined by natural features and/or invisible gov-mandated lines. |

Census Block | A section of a census tract and smallest unit of area for which the Census Bureau gathers 100% data. In urban areas, census blocks coincide with with city blocks and defined by streets. In rural areas, may be much larger. 8 million census blocks in US. |

Census block groups | Census block groups consist of a group of census blocks, and are the smallest unit of area used to gather sample data. |

Census designated place | aka CDP, is a densely populated area that has a name designation but is not part of an incorporated place. Wit hthe help of state and local governments, the Census Bureau identifies CDPs for incorporation into censuses. |

place | A densely populated area, and can be classified as either an incorporated place or a CDP. |

Urban cluster | A densely populated area containing between 2,500 and 50,000 people. |

Urbanized area | A minimum population density of 1,000 every square mile, and a total population of at least 50,000 people. It must include a minimum of one central place with a surrounding territory. |

Metropolitan statistical area | MSA has a densely populated urban core, upon which surrounding communities are socially and economically dependent. MSAs often cover multiple counties. Core area contains either 50,000 people living in a city or 100,000 people living in an urbanized area |

New England MSAs | In New England, the amount is reduced to 75,000 and MSAs are defined by civil divisions, population density, and commuting patterns. |

Primary metropolitan statistical area | PMSA, may cover multiple counties if those counties are connected by large amounts of commuting. An MSA with 1,000,000 or more people may include multiple PMSAs |

Consolidate metropolitan statistical area | CMSA, is created when two or more PMSAs are located within the same geographic area. To be classified as a CMSA, a region must have a population totaling 1,000,000 people, consist of multiple PMSAs, and qualify to be an MSA. |

Economic Census | Tallies the number of businesses in: manufacturing, construction, minority-owned, female-owned, mining and quarrying, service, transportation, retail trade, and wholesale trade. The Census Bureau takes economic census for every year ending in a 2 or a 7. |

North American Industry Classification System | NAICS - created by experts in the US, Canada, and Mexico and replaced the existing classification systems within each country. US used to use the "standard industrial classification system" (SIC). |

NAICS breakdown | Industries are represented between 2 and 6 digits depending on level of detail. Sectors, which = broadest industry classifications, have two digits. Individual sectors, which are the most specific industry classifications, have 6 digits. |

TIGER database | TIGER (topologically integrated geographic encoding and referencing system) database was created by the US Census Bureau to fulfill geographic functions for the 1990 census. |

TIGER database (cont.) | Duties included: generating cartographic products, creating a geographic structure that gathers and send out statistical info, matching the appropriate addresses with the appropriate geographic locations, creating a municipall GIS. |

ZIP code | "zone improvement plan" codes designate a street section, group of streets, buildings, or a set of P.O. Boxes for purposes of mail delivery. They were created by the USPS |

ZCTA | Zip Code Tabulation Area was created by the US Census Bureau to define more specifically the geographic area designated by ZIP codes. ZCTA was used during the 2000 Census to generate summary statistics. |