UConn HDFS 2004W Exam 2

Purpose of Sampling

It is usually not feasible to study an entire population due to:

-Size

-Availability

-Time

-Cost

-Sampling allows for representation of population

Representativeness

Quality of a sample having the same distribution of characteristics as the population from which it was selected

Probability Sampling

-Used when researchers want precise, statistical descriptions of large populations

-A sample of individuals from a population must contain the same variations that exist in the population

Random Selection

Each element has an equal chance of selection independent of any other event in the selection process

Population

Aggregation of elements from which the sample is selected

Element

Unit about which information is collected that provides basis of analysis

-the participant = an element

Sampling Frame

That list or quasi list of units composing a population from which a sample is selected

-Example: rosters, phone book, etc.

Sampling Unit

Element or set of elements considered for selection in some stage of sampling

-The sample you end up with

Parameter

Summary description of a given variable in a population (describes the population)

Statistic

Summary description of a variable in a sample (describes the sample)

Sampling error

Sampling error is the error caused by observing a sample instead of the whole population

-as sampling size increases, sampling error decreases

Confidence level

The estimated probability that a population parameter lies within a given confidence interval

confidence interval

The range of values within which a population parameter is estimated to lie

Types of Probability Sampling Designs

Simple random sampling (SRS)

Systematic sampling

Stratified sampling

Simple Random Sampling (SRS)

-Feasible only with the simplest sampling frame.

-Sampling units are assigned numbers and then randomly selected.

-Not the most efficient method if done manually

Systematic Sampling

Random starting point and select that unit and every subsequent unit at interval k

Sampling ratio and interval

-Proportion of elements in the population that are selected

-List of 100 names - you want to select 20.

Sampling ratio would be 20/100 or 1:5.

Sampling interval would be 100/20 or 5

Stratified Sampling

Grouping of units composing a population into homogenous groups before sampling

-improves the representativeness of a sample

Nonprobability Sampling

Technique in which samples are selected in a way that is not suggested by probability theory

Types of Nonprobability Sampling

-Reliance on available subjects

-Purposive or judgmental sampling

-Snowball sampling

-Quota sampling

Reliance on available subjects

-Only justified if less risky sampling methods are not possible.

-Researchers must exercise caution in generalizing from their data when this method is used

Purposive or judgmental sampling

Selecting a sample based on knowledge of a population, its elements, and the purpose of the study

Snowball sampling

-Appropriate when members of a population are difficult to locate.

-Researcher collects data on members of the target population he/she can locate, then asks them to help locate other members of that population

Example: homeless people

Quota sampling

-Begin with a matrix of the population.

-Data is collected from people with the characteristics of a given cell.

-Data should represent the total population

Example: differences in sex

Goal of social science

To promote human welfare

Impact of psychological/social research

-Health

-Law and criminal justice

Generalization

-The extent to which a single study conducted with a specific sample and procedure can be generalized to other populations

-With enough evidence we can make predictions about human behavior

External validity

the extent to which findings can be generalized

Issues related to generalization

-Use of college students

-Gender considerations

-Volunteers

-Geographic location

-Culture

Use of college students / generalization

-Possess characteristics of developing young adults

-Intelligent, high cognitive skills, know how to win approval from authority

-A sense of self-identity

-Social and political attitudes

-Need for peer approval

-Unstable peer relationships

Gender considerations / generalization

-Rely on either males or females because of convenience. Gender bias

volunteers / generalization

-tend to be highly educated

-more social

-Higher SES

-Need for approval

-Healthy

Evaluating generalizations

-Exact Replications

-Conceptual replications

-Lit reviews

-Meta analysis

Exact Replications

-An attempt to replicate precisely the procedures of a study

-Make sure whether the same results are obtained with replication

Conceptual Replications

-The use of different procedures to replicate a research finding

-The IV from previous study is manipulated in different ways

Utilizing lit reviews for evaluating generalizations

-Summarizes what has been found

-Tells the reader:

--What findings are strongly supported

--What findings are weakly supported

-Exposes inconsistent findings and areas lacking proper research

-Discusses future directions for research

Meta analysis

method for determining the reliability of a finding by examining the results from many different studies

Quantitative research helps us to:

-Better understand phenomenon and gain new perspectives

-Collect and explore in-depth information that can't be conveyed quantitatively

-Provide better descriptions of complex phenomena

-Explore sensitive topics

Types of quantitative approaches

-Narrative

-Phenomenology

-Grounded Theory

-Ethnography

-Case Study

Narrative

to describe individual stories arranged in chronological order

Phenomenology

to describe a shared phenomenon through commonalities in many individual stories

Grounded theory

To generate or discover a theory

Ethnography

to describe and interpret the shared patterns, behaviors, values, and language of a culture sharing group

Case study

to study an issue explored through one or more cases within a bounded system (i.e. setting, context)

How to get participants:

-gaining access and interest within participant groups

-Stakeholders and gatekeepers

-Introductory phase

-Engaging and keeping interest

-Incentivizing

reliance on available subjects (convenience sampling)

-Sampling from subjects who are available

-only justified if less risky sampling methods are not possible

-Hard to generalize

Purposive or Judgmental Sampling

-selecting a sample based on knowledge of a population, its elements, and the purpose of the study

Snowball sampling

-appropriate when members of a population are difficult to locate

-asks participants to recommend other participants/help locate them

Quota sampling (non-probability)

-being with a matrix of the population

-data is collected from people with the characteristics of a given cell

-data should represent the total population

Experiments involve:

-Manipulation of a variable and observing the consequences of that action

-Straightforward manipulation

-Staged manipulations

Components of a classic experiment

-Experimental and control group

-Randomization or matching

-Pre-test - post-test design

Pre-test - post-test design

Capturing data prior to intervention and after the intervention

Confounding variable

An uncontrolled variable that might be responsible for the observed effect on an outcome

Ex) ice cream consumption and murder rates

Independent group's design

Different participants assigned to each of the conditions using random assignment

Repeated measures design

The same individuals will participate in both conditions. (e.g., placebo pill vs. actual treatment)

Probability Sampling

The degree of representativeness achieved is a function of sample size

double-blind experiment

An experimental design in which neither the subjects nor the experimenter know if the placebo will be administered or not

Type of quasi experiments

-one-shot case study

-one-group pretest-post test design

-static-group comparison

one-shot case study

-single group is studied at a single point in time after some treatment that is presumed to have caused change

one-group pretest-posttest design

single group is observed at two time points, one before the treatment and one after the treatment

Static-group comparison

a group that has experienced some treatment is compared with one that has not Observed differences between the two groups are assumed to be the result of the treatment

What is collected via surveys?

-attitudes and beliefs

-facts and demographics

-behaviors

Double-barreled questions

asking two questions in one

Loaded questions

They are questions which contain emotive language which is likely to produce an emotional reaction or desired response from the respondent

negative wording

avoid phrasing questions with negatives such as not or don't

What does avoid jargon mean?

avoid words that people will not understand

guidelines for questionnaire construction

One question per line.

Use contingency questions when necessary.

Format questions so they are easily answered.

Response rate

Number of people participating in a survey divided by the number selected in the sample

Acceptable response rates + why is it important?

50% - adequate for analysis and reporting

60% - good

70% - very good

*Important for generalizability or sample representativeness

Types of survey administration

self-administration,

face-to-face interview, telephone,

new technologies

strengths of survey research

-Useful in describing the characteristics of a large population.

-Make large samples feasible.

-Flexible - many questions can be asked on a given topic

Weaknesses of Survey Research

-Can seldom deal with the context of social life.

-Inflexible in some ways. -Cannot be modified once administered.

--Subject to artificiality

Secondary analysis

A form of research in which the data collected and processed by one researcher are reanalyzed—often for a different purpose—by another

Why conduct evaluation research?

-To determine whether a social intervention or program has produced the intended result

- program regulation

Types of evaluation research

-program evaluation

-needs assessment studies

-cost-benefit studies

-monitoring studies

Program evaluation

The determination of whether a social intervention is producing the intended result.

Needs assessment studies

Studies that aim to determine the existence and extent of problems, typically among a segment of the population, such as the elderly or at-risk adolescents.

Cost-benefit studies

Studies that determine whether the results of a program can be justified by its expense (both financial and other)

Monitoring studies

Studies that provide a steady flow of information about something of interest, such as crime rates or the outbreak of an epidemic

Experimental context

aspects of the context of an experiment that might affect the experiment

Types of evaluation research designs

-experimental designs

-quasi-experimental designs

-social indicators research

time series deisgn

studies that involve measurements taken over time

non-equivalent control group

A control group that is similar to the experimental group but is not created by the random assignment of subjects

Multiple Time-Series Designs

-An improved version of the nonequivalent control group design

-Allows multiple assessments of multiple comparisons over time

Why results are ignored?

-Implications may not be presented in a way that nonresearchers can understand.

-Results sometimes contradict deeply held beliefs.

-Vested interest in a program.

Social indicators research

-Combines evaluation research with analysis of existing data

-Provides an understanding of broader social processes

-Example: social indicators help determine whether the death penalty works as a deterrent

Codebook

Document that describes the locations of variables and lists the assignments of codes to the attributes composing those variables

Univariate

simplest form, describe a sample in terms of a single variable

Used to DESCRIBE

Bivariate

describe a sample in terms of two variables simultaneously

Used to EXPLAIN

Multivariate

analysis of two or more variables simultaneously

Used to EXPLAIN

Frequency distribution

Description of the number of times that the various attributes of a variable are observed in a sample

Dispersion

Refers to the way values are distributed around some central value

Range

the distance separating the highest value from the lowest value

standard deviation

-Index of the amount of variability in a set of data.

-A higher standard deviation means that the data are more dispersed; a lower standard deviation means that they are more bunched together

correlation coefficient

A statistic that describes the strength and direction of a relationship between two variables

Strength of effect

Small = .20

Medium = .30

Large = >.40