Differences between revisions 1 and 2
⇤ ← Revision 1 as of 2020-10-22 17:24:44 →
Size: 2353
Editor: DominicRicottone
Comment:
← Revision 2 as of 2020-10-22 17:37:02 → ⇥
Size: 2509
Editor: DominicRicottone
Comment:
Deletions are marked like this. Additions are marked like this.
Line 2: Line 2:

There are a number of considerations for analyzing survey data.
Line 11: Line 13:
There are key differences between surey statistics and model statistics. There are key differences between the statistics employed in models and in surveys.
Line 16: Line 18:
Build a mathematical model that describes a population.
Generate a random sample from that population to generate estimates.
Estimate how the error terms of those estimates would vary if repeated samples were drawn.		  1. Build a mathematical model that describes a population.
 2. Generate a random sample from that population to generate estimates.
 3. Estimate how the error terms of those estimates would vary if repeated samples were drawn.

In other words, the key is how well the model describes the population.
Line 23: Line 27:
Identify a population with fixed descriptives.
Randomly draw a sample from that population to collect measures from.
Estimate how the measures would vary if repeated samples were drawn.		  1. Identify a population with fixed descriptives.
 2. Draw a sample from that population to collect measures from.
 3. Estimate how the measures would vary if repeated samples were drawn.

In other words, the key is how well the sample fits the population. If the full population were contacted, measures would be perfect.
Line 30: Line 36:
Using model-based inference while accounting for survey design Using model-based inference while accounting for survey design.
Line 38: Line 44:
The population of a survey has fixed but unknown descriptives.
A random sample is contacted, rather than the full population, for reasons of cost and administration.
The population descriptives are estimated based on the design of the sample and the descriptives collected from that sample.		 A sample is contacted, rather than the full population, for reasons of cost and administration.
There are costs paid for this decision.
Line 74: Line 79:


Survey Statistics

There are a number of considerations for analyzing survey data.

Contents

  1. Survey Statistics
    1. Statistics with Survey Data
      1. Model-based inference
      2. Design-based inference
      3. Inferential statistics from complex survey data
    2. Survey Populations
      1. Limitations of Survey Sampling
      2. Limitations of Surveying
      3. Common Sampling Frames

Statistics with Survey Data

There are key differences between the statistics employed in models and in surveys.

Model-based inference

  1. Build a mathematical model that describes a population.
  2. Generate a random sample from that population to generate estimates.
  3. Estimate how the error terms of those estimates would vary if repeated samples were drawn.

In other words, the key is how well the model describes the population.

Design-based inference

  1. Identify a population with fixed descriptives.
  2. Draw a sample from that population to collect measures from.
  3. Estimate how the measures would vary if repeated samples were drawn.

In other words, the key is how well the sample fits the population. If the full population were contacted, measures would be perfect.

Inferential statistics from complex survey data

Using model-based inference while accounting for survey design.

Survey Populations

A sample is contacted, rather than the full population, for reasons of cost and administration. There are costs paid for this decision.

Limitations of Survey Sampling

If a sample is a poor fit for the population, then it will be difficult/impossible to estimate population descriptives. This is why samples are *randomly* drawn.

But for random sampling to succeed, the target populations needs to be completely identified. Incompleteness of data can skew samples.

Limitations of Surveying

Non-random non-response is an additional roadblock to estimating population descriptives.

Some populations are inherently difficult to contact, due to:

  • political, legal, and ethical considerations (i.e. minors in the population)
  • language and cultural barriers + remoteness of location (i.e. rural Alaska)

The availability of contact data often dictates the mode of survey instrument. Some populations are not easily contacted by specific modes, due to:

  • socioeconomic conditions (i.e. lack of telephone connectivity)
  • low accuracy contact data (i.e. outdated addresses)

Common Sampling Frames

  • Census Bureau surveys are excellent sources of regional, hierarchical data (i.e. states > counties > tracts)

  • U.S. Postal Service Delivery Sequence files
  • Random digit dialing

CategoryRicottone

Statistics/SurveyInference (last edited 2025-01-10 15:49:44 by DominicRicottone)