Week 1 notes
Week 1 notes SWRK 344
Popular in Social Work Statistics and Data Analysis
verified elite notetaker
Popular in Social Work
verified elite notetaker
This 5 page Class Notes was uploaded by Kirsten Swikert on Tuesday September 6, 2016. The Class Notes belongs to SWRK 344 at Western Kentucky University taught by Dr. Getch in Fall 2016. Since its upload, it has received 6 views. For similar materials see Social Work Statistics and Data Analysis in Social Work at Western Kentucky University.
Reviews for Week 1 notes
Report this Material
What is Karma?
Karma is the currency of StudySoup.
You can buy or earn more Karma at anytime and redeem it for class notes, study guides, flashcards, and more!
Date Created: 09/06/16
Statistics: collecting, analyzing, interpreting, presenting, and organizing data; using data for practice Why is it important? • So we can do our own research and understand others • Terminology • Helps with observations so we can better help others Data: • Measurements that are collected for a specific research purpose • Data exists before analyses are run • Data can be qualitative, quantitative, etc. • The following methods can be used to collect date: o Observation (direct, participant) o Telephone, mail, electronic surveys o Interviews (group, individual, focus groups) o Content analysis or other analysis o Historical research o Community needs assessment Primary and Secondary data analysis • Primary data analyses includes the analysis of data collected for a specific purpose • Secondary data analyses includes previously collected data o May have been collected for another purpose o May have been previously analyzed o May be stored in an accessible database Data sets: • Data gathered in research studies that are collected for a specific purpose • There can be multiple data sets within one research study • Data sets are useful but often are limited as it is impossible to collect data on every possible variable or constant • It is necessary to determine what data is most important to collect and to use time and other resources wisely o This is why researchers will often conduct a review of the literature before beginning a study o Collecting erroneous data can discourage participation o Failing to collect important, necessary data can undermine an important research study Variables and Constants: • Variables are traits or characteristics that differ in quality or quantity, they can be measured or observed o Age, gender, education, income • Constants are traits or characteristics that do not vary in quality or quantity, they remain constant o Once assigned a number, it remains constant (x-6=3 3 and 6 are constants) o In a data set, a constant could be a football team, the variables could be the height/weight of each player Attributes: • Categories are assumed by a variable, these categories can be expressed by words or numbers o Those words and numbers are referred to as attributes • Value labels are attributes that are expressed by words • If the attribute is expressed in numbers, it is commonly referred to as the value or values • Attributes describe the variable in terms of words or numbers o Tall, short, average o 6’4”; 4’11”; 5’5” Frequency: • Is simply the number of times an attribute actually occurs within a specified data set • Remember variables vary, so in one sample or data set a variable might occur 15 times and in another the frequency of that same variable might be 3 • If the variable does not change, then it is a constant • What are the following? o Type of band/musical instrument played: value label o Flute, trumpet, clarinet, trombone, and saxophone: values o 8, 12, 13, 9, and 5: quantitative o Woodwind, brass, percussion, strings: qualitative Conceptualization and Operationalization • To design a research project, you need to know what you want to study and what you want to find out, this process is called conceptualizations o Two steps: § Selecting variables of interest, these should be the most important variables to study § Concisely describing each variable, we cannot measure a variable if it is not concisely described • Operationalization is when we specify how we are going to measure the variables o What are indicators? What are we measuring and how are we measuring it? Measurement and Measurement Issues • Data can be provided by people and this can create issues if there are not specified instructions for how the data is to be collected. • Multiple indicators are often used to measure a variable. Multiple indicators allow us to gain a better understanding of how a variable is impacted by a phenomenon. Reliability: • Consistency • Does the measurement produce similar (consistent) results over repeated measurements • Measurements vary in their reliability • There is not “perfect” reliability • We are looking for consistent measurement Validity: • Does the instrument measure what its purports to measure • Is it biased? • Does it “concur” with other instruments? • Basically does it measure what we expect it to measure o Scales o Thermometer o Measuring tape o Exams Validity and reliability are necessary: • A reliable measure that is not valid will not provide you with the data needed because you will not be measuring the phenomenon you wish to measure • A valid measure that is not reliable means your results will vary and you will not be able to tell if the intervention made a difference or not • Most research studies will report several kinds of validity o Face validity o Content validity o Concurrent validity o Predictive validity o Construct validity Measurement and measurement precision: • Precision allows accuracy of results and allows one to have confidence in the findings • There are four levels of precision in measurement o Nominal o Ordinal o Interval o Ratio • Nominal is the lowest level of precision and ratio has the highest level of precision, however, it is difficult to find measurements at the ratio level Nominal level: • Variable categorized into discrete attributes • It represents a difference in kind • It does not describe the quantity of the variable being measured • You cannot rank order the variables attributes • It must consist of two or more attributes (otherwise it could be a constant) • For example: o Yes, no, indifferent, undecided o Vanilla, strawberry, chocolate o Republican, democrat, independent Ordinal level: • It has more than one attribute • Variable categorized into discrete attributes • It represents a difference in kind • You can rank order because it has a quantitative meaning • The intervals between attributes cannot be assumed to be the same o We can’t say that an attribute is twice as much or that the difference between the attributes is exactly the same • Different people would rank order the same o High, medium, low o Small, medium, large Interval level: • It has more than one attribute • Variable categorized into discrete attributes • It represents a difference in kind • You can rank order because it has a quantitative meaning • The intervals (distance) between attributes can be assumed to be the same meaning that one attribute can double or triple another attribute o 1 year, 2 years, 3 years o Temperature can be ½ as cold or twice as hot o You can specify the units • It does not have an absolute zero Ratio level: • It has more than one attribute • Variable categorized into discrete attributes • It represents a difference in kind • You can rank order because it has a quantitative meaning • The intervals (distance) between attributes can be assumed to be the same meaning that one attribute can double or triple another attribute • It has an absolute zero o The zero is fixed, not arbitrary, and indicates absence of the phenomenon (0 children) o Specifies the amount of the property being measured o We can add, subtract, multiply, and divide the amounts with accuracy because we cannot have negative values because there is an absolute zero Interval/Ratio: • To be interval or ratio level the underlying variable must have measurement precision o If in doubt, go with the lower level of measurement § If it appears interval but ordinal is a more accurate description, use ordinal o In social work practice, the measurement of a skill may lack measurement precision § Other factors may confound our conclusion § There may be several variables that are responsible for the “skill” • Absolute zero: ratio; Negative number: interval
Are you sure you want to buy this material for
You're already Subscribed!
Looks like you've already subscribed to StudySoup, you won't need to purchase another subscription to get this material. To access this material simply click 'View Full Document'