Embed Code - If you would like this activity on your web page, copy the script below and paste it into your web page.

Normal Size Small Size show me how

Normal Size Small Size show me how

# SW R & STATS Midterm

Hypothesis | proposes a relationship between 2 or more variables. One variable influences another variable. |

Independent Variable | the variable that causes or leads to another variable. |

Dependent Variable | this variable varies depending on the other variable. |

Concept | is a mental image that summarizes a set of similar observations, feelings, or ideas. |

Conceptualization | is the process of specifying what we mean by a term. |

Operationalization | is the process of connecting concepts to observations. When we identify specific observations that we will take to indicate that concept in empirical reality. |

Research Questions emerge from: | Own experiences, Research literature; Theories; Pragmatic(real world) |

We evaluate research on: | Feasibility; Social Importance; and Scientific relevance |

What is sampling? | -some part of a larger body specifically selected to represent the whole. It is taking any portion of a population or universe as representative of the population or universe. |

Sampling | The procedure by which a few subjects are chosen from the universe to be studied in such a way that the sample can be used to estimate the same characteristics in the total. |

Advantages of Sampling | Less costly, quicker and, if selected properly, gives results with known accuracy that can be calculated mathematically. |

Why use Sampling? | Impossible to observe all relevant events. Time Cost |

Population | The collection of all individuals, families, groups, organizations, and events that we are interested in finding out about. |

Target Population | Is the population to which the researcher would like to generalize his or her results. |

Survey Population | Operational definition of the target population; that is, a target population with explicit exclusions. |

Element/Unit of Analysis | The unit about which information is collected and that provides the basis of analysis. Each member of a population is an element. |

Sample Design | A set of rules or procedures that specify how a sample is to be selected. This can be either probability or nonprobability. |

Sample Size | The number of elements in the obtained sample. |

Sampling Error | The degree of error to be expected for a given sample design; the difference between the sample mean and the population mean. |

Sampling Bias | The notion that those selected are not "typical" or "representative" of the larger populations that have been chosen from. |

Confidence Level | -How often one could expect to find similar results if the survey were repeated. -The degree of certainty of obtaining the same results. -Often informs about how often the findings will fall outside the margin of error. |

Confidence Interval | The range in which we are fairly certain that the population value lies. |

Statistical Inference | The process of reasoning by which information about a population is extracted from a sample data. |

Probability Sampling | {known in advance] every person in the population has an equal chance of being selected. Each unit in the population has an equal chance of being selected. |

Non-Probability Sampling | we do not know [in advance] if every person in the population has an equal chance of being selected. Chance of being selected in unknown. |

Types of Probability Sampling | -Random Sampling -Systematic Sampling -Stratified Sampling -Cluster Sampling |

Random Sampling [Probability Sampling] | -selected by chance. -Must have equal access to everyone in the population at the same time. -Lottery |

Systematic Sampling [Probability Sampling] | -Choice of being selected in a sample in not random, but systematic choice. -Ex: First element is selected randomly then every "nth" participant is selected. |

Stratified Sampling | -stratifies the population into subgroups and then randomly selects from each subgroup. -All elements in the sampling frame are distinguished according to their value on some relevant characteristic. -that characteristic forms the sampling STRATA |

Cluster Sampling | -used when people in a population of interest cannot be identified in a direct way. -Less likely to be representative because more than one stage is needed to identify the sample. Each stage poses likelihood of error in selection. |

Types of Non-Probability Sampling | -Convenience Sampling -Quota Sampling -Criterion Sampling -Snowball Sampling |

Convenience Sampling [Non-Probability] | Selection of elements is done by what is available and/or convenient to the researcher. |

Quota Sampling [Non-Probability] | -intentionally compare 2 or more subgroups. -does not use random sampling -pre-set number of elements based on characteristics in a population to ensure that the sample represents those characteristics in the population. |

Criterion Sampling [Non-Probability] | -selects participants based on a set of criteria related to the purpose of the study. -AKA Purposive Sampling |

Snowball Sampling [Non-Probability] | -used when it is difficult to identify or locate the kinds of people who are the focus of the study. -select and interview people who fit criteria and then participants are asked to identify and recruit others. |

How to determine sample size: | -Type of analysis to be employed -The level of precision needed -Population homogeneity/heterogeneity -Available resources -sampling techniques used in research with population. |

Sample size in types of sampling: | Probability: samples are larger because they have different purposes. They need to mathematically account for or measure error. Non-Prob: is usually used in exploratory studies when we need to know more about the population. |

Non-Sampling Errors | -an inadequate sampling frame -nonresponse from participants -field errors -response errors coding and data entry errors |

Factors that affect choice of sample design: | -stage of research -data use -available resources for drawing sample -nature of the research design |

Quantitative Measures | When we have numbers we can analyze information using statistics. -Ex: height, weight, how many, pounds, test scores, income. |

Qualitative Measures | We can use open ended questions to get sentences, paragraphs etc. -Ex: gather words, gender, race, religion, satisfactions. -we gain flavor, attitudes, intentions from participants, but we have more difficulty using statistics. |

Levels of Measurement | -Nominal -Ordinal -Interval -Ratio |

Nominal | -With this measurement, you indentify variables whose values have no mathematical interpretation; they vary in kind or quality but not in amount. -Qualitative, no mathematical interpretation. |

Ordinal | With this measurement, you specify only the order of the cases in "greater than" and "less than" distinctions. |

Interval | With this measurement, numbers are represented in fixed measurement units but have no absolute zero point. |

Ratio | With this measurement, numbers are represented in fixed measuring units with an absolute zero point. -can be added and subtracted as well as multiplied and divided. |

How can we get data? | -Observations -Use existing data -Use surveys/questionnaires -Triangulation: combine measure-most useful |

Direct Observation | -Observing someone/thing directly |

Indirect Observations | -Unobtrusive measures: information is gathered without direct knowledge or participation. -Content Analysis: looks at representations of the topic in various media forms such as new reports, tv shows, magazines, etc. |

Existing Data | -Archived data set at the federal and state level or at the agency level. -Certain types of annual or less frequent reports, i.e, the census. |

Surveys/Questionnaires | -Self-report and administered -Standardized -Can test a variety of problems for many ages -Many surveys can be used that are already developed-convenient. |

Scales and Indexes | -A group of questions in which the total responses to the questions are summed or in some other way manipulated to provide a more complex or complete measure of a concept than can any single question or component element of the scale or index. |

Reliability | -is concerned with questions of stability and consistency. Does the same measurement tool yield stable and consistent results when repeated over time. -refers to a condition where a measurement process yields consistent scores over repeated measurements |

Validity | refers to the extent we are measuring what we hope to measure(and what we think we are measuring). |

Types of Reliability | -Test Retest -Inter-item -Interobserver |

Test-Retest Reliability | When the researcher administers the same measurement tool multiple times, asks the same question, follows the same research procedures, etc. -Does she obtain consistent results? -Simplest method for testing reliability. |

Inter-item Reliability | -this is a dimension that applies to cases where multiple items are used to measure a single concept. |

Interobserver | -is concerned to the extent to which different interviewers or observers use the same measure and get equivalent results. -if different observers or interviewers use the same instrument to score the same thing , their scores should match. |

Face Validity | this criterion is an assessment of whether a measure appears , on the face of it, to measure the concept it is intended to measure. -very minimal assessment. -if a measure cannot satisfy this criterion, then the other criteria are inconsequential. |

Types of Validity | -Face Validity -Content -Criterion-related -Construct |

Content Validity | -is concerned with the extent to which a measure adequately represents all facets of concept. |

Criterion-related Validity | -applies to instruments that have been developed for usefulness as indicator of specific trait or behavior, either now or in the future. -Ex: an individual's performance on a driving test correlates well with his/her driving ability. |

Construct Validity | -is concerned with the extent to which a measure is related to other measures as specified by theory or previous research. -is established if the measure covers the full range of a concepts meaning. |

Criterion Validity | -Predictive: which is the ability of one measure to predict the score on the criterion measure in the future. -Concurrent: when the core on the new measure resembles in some predefined fashion the scores on the criterion measure. |

Five Criteria for establishing a casual relationship: | -Empirical association -Temporal priority of the independent variable -Non-spuriousness -Identifying a causal mechanism -Specify the context in which the effect occurs |

Empirical Association | The independent variable and the dependent variable must vary together. -A change in X is ASSOCIATED with a change in Y. |

Temporal Priority of the Independent Variable | The change in X must occur before the change in Y. |

Non-spuriousness | We say that a relationship between two variables is spurious when it is due to variation in a third variable; so what appears to be a direct connection is in fact not. |

Casual Mechanism | Is the process that creates the connection between the variation in an independent variable and the variation in the dependent variable it is hypothesized to cause. |

Context in which the effect occurs | No cause has its effect apart from some larger context involving other variables. -When, for whom, and in what conditions does this affect occur? -a cause is really one among a set of interrelated factors required for the effect. |

Experimental Group | -is the group of subjects in an experiment that receives the treatment or experimental manipulation. |

Comparison Group | -is the group of subjects that is exposed to a different treatment than the experimental group(or that has a different value on the independent variable). |

Control Group | -is the group of subjects that receives no treatment instead of a different treatment. |

PreTest | is a measurement of the dependent variable prior to initiating the treatment(independent variable). |

Post Test | is a measurement of the dependent variable subsequent to the treatment(independent variable). |

Internal Validity | -whether the intervention rather than other factors is responsible for improvement. What other factors may be responsible for change? |

External Validity | -Can the results of this sample be generalized to the larger population? |

Research Design TYPES | -Pre-Experimental -Quasi-Experimental -True Experimental |

Pre-Experimental Design | 1.One-group Post test only design: an intervention followed by a measure, weakest, cross-sectional. 2.One-group Pretest/post test design: pretest, then intervention, then post test, a little stronger, longitudinal. |

Quasi-Experimental Design | 1. Pretest/Post test Comparison group design: has comparison group which does not receive intervention, longitudinal. 2. Time series design: multiple measures of the client outcome prior and after intervention, no comparison or control group, longitud. |

True Experimental Design | 1. Prestest/Post test Control group design: pretest/post test design that involves a control group, longitudinal. RANDOM. |

Two Ethical Issues | Deception:in general subjects must be informed as the the purpose of the study. Distribution of Benefits: to what degree are persons harmed who do not receive the benefit of the treatment or who received the treatment when compared to the other group. |

Independent Variable | -the variable that causes or leads to another variable. -in intervention research, this can be the treatment variable. |

Dependent Variable | -this variable varies depending on the other variable. -in the intervention research, this can be the outcome variable. |

Mode | -the value that occurs most often in a series of numbers. |

Median | -middle value when numbers are arranged in order. (better to use median than the mean when there are outliers) |

Mean | -Average: sum of all the numbers divided by the total numbers. |

Frequencies(measures) | -percentages help us visualize the data in charts and graphs. |

Valid Percent(measures) | -is the number of the item including the missing or other codes. |

Cumulative Percent(measures) | -is the number of ind's in that grouping and below. |

Varience | -average squared deviation of the mean. |

Standard Deviation | -the amount, on average, that scores or responses vary from the MEAN score. |

Inferential Statistics | -refers to the various tools that help us determine how much confidence we have when we generalize our findings from a sample to a population. -in this case, we always assume that there will be some "margin or error". |

Hypothesis Testing | Null Hypothesis: there is no relationship/association between variables. Alternative Hypothesis: there is a relationship between the variables. |

Errors | Type 1 Error: you reject the NULL and it is true. Type 2 Error: you accept the NULL and it is false. |

Statistical Significance | -relationship between two variables that is determined by math principles--based on a probability that making this claim, I will only be wrong less than .05 percent of the time. I am 95% confident that my answer is not due to error. |

Clinical Significance | -is determined by the judgement of professional--exploring the claim that a relationship exists between an intervention and a client outcome variable. -single subject designs help do this. |

Cross-Tabs Contingency Tables | -builds upon our understanding of the Bernoulli Process. -used for categorical data to examine if two variables are independent of each other or is there an association between them. -this independence may be tested using a chi-square statistic. |

Bivariate Tests | Chi-Square: determines whether 2 variables at the nominal or ordinal levels are statistically independent of each other. -used for categorical variables |

Student's t-test types | Independent Sample: this means from 2 different groups of people. Paired Sample: this means fro the same group of people measured twice. -SPSS has both commands for these. |

T-test's assume... | -outcome variable is normally distributed. -sample is randomly selected. -sample size may be small. |

Correlation Coefficient | -how 2 continuous variables may be related to each other. -tells the strength of a relationship between 2 variables and also its nature. -values from -1.0 to 1.0 |

Regression Analysis | -this technique is used to PREDICT the value of the dependant variable(outcome, y) from one or more independent variables(x) -this prediction is usually in the form of a linear equation. |

Multiple and Logistic Regression | Multiple Regression:allows you to use 2 or more independent variables to predict a continuous dependent or outcome variable. Logistic Regression: is used for predicting a dichotomous outcome variable. |

Research Questions | -questions about the social world that you seek to answer through the collection and analysis of firsthand, verifiable, empirical data. |

Research Foundations | -empirical or actual research studies -single case designs -Lit. Reviews -books and book reviews |

Theories | -help us make sense of social phenomena. -they are used to explain and predict behavior/attitudes etc. |

Scientific Method's findings should be.... | true, repeatable, generalizable, and it's knowledge is valid and reliable. |

Social Science Approach | -relies on logical and systematic methods to answer questions and its done so others can evaluate, dispute or replicate the findings. -asking questions, observing, counting, are all basic methods used. |

Types of Research | -Descriptive -Exploratory -Explanatory -Evaluation |

Descriptive Research | -defining and describing social phenomena, often first done on a new topic. |

Exploratory Research | -no explicit expectations, captures large amounts of unstructured info. |

Explanatory Research | -identifies cause and effect and to predict how one phenomenon will change or vary in response to another phenomenon. |

Evaluation Research | -determines the effects of a program or intervention. |

Quantitative methods | -surveys or experiments which produce numbers, percentages, counts. |

Qualitative methods | -interviews, focus groups, surveys which produce written or verbal comments or text. |

Triangulation | -use of multiple methods of Quantitative and Qualitative. |

Errors in Research | -Observing: selective observation and inaccurate observation. -Generalizing: over generalization -Reasoning: jumping to conclusions. -Reevaluating: resistance to change(reluctance to modify ideas in light of new info) |

Goal of Research | to figure out how or why the social world operates as it does. It is the goal of validity---truth. |

Research | -is defined as a systematic investigation, including research development, testing and evaluation, designed to develop or contribute to generalizable knowledge. |

Human Subject | -is defined as a living individual about whom an investigator conducting research obtains (1) data through intervention or interaction with the individual, or (2) identifiable private information. |

Ethical Problems in Research | -Physical Harm -Psychological Harm -Invasion of Privacy (without consent) -Deception -Misrepresentation of Findings -Balancing risks and gains |

Confidentiality | -the researcher knows the names of the participants but promises not to disclose those names to anyone outside of the team. Data cannot identify anyone. |

Anonymity | the researcher does not know the names of participants. |

Informed Consent | the researcher must give info to potential participant about the study and risks and benefits of it. Participant can stop at any time. |

Institutional Review Board | group at an institution who review potential studies and determine if they are ethically sound. Must approve before beginning. |

Created by:
TLynn1983