Class Notes (1,100,000)

CA (630,000)

U of G (30,000)

SOAN (900)

SOAN 2120 (400)

David Walters (100)

Lecture 10

# SOAN 2120 Lecture Notes - Lecture 10: Sampling Distribution, Sample Size Determination, Microsoft Powerpoint

by OC1223415

School

University of GuelphDepartment

Sociology and AnthropologyCourse Code

SOAN 2120Professor

David WaltersLecture

10This

**preview**shows page 1. to view the full**5 pages of the document.**SOAN 2120 – Introductory Methods – October 3rd, 2017

Read Chapter 5

Chapter 5

• NVivo 11 – a qualitative software program that is designed to facilitate qualitative research

o This software consolidates and collates the various data sources that a researcher may

have by electronically organizing them and allowing for retrieval of items from locations

• Theory emerges from the research process rather than being the central

framework that informs and directs the research project from the beginning

• The employment of qualitative software should never substitute a deep personal understanding

• NVIVO Program

o Sources – contains the data sources that you import into your projects

▪ Internals

• You can create subfolders to organize your material in

▪ Externals

• Contain references and links to source material that cannot be directly

imported into your project (ex: hardcopy books, webpages, PowerPoint)

▪ Memos

• A site where you can note your reflections, thoughts, ideas, questions

• Nodes – a receptacle for all references to a particular theme or concept

o Picture it as a box with thematic label: each time you code something into a theme

• NVivo enhances the eseahe’s ability to organize, link and ask questions of the data with goals

• Concerns with NVivo

o Computer-assisted research remains true to the spirit of qualitative methods

o Use of such software leads to a greater separation between data/investigator

o Loss of contextualizing detail when field notes or media sources are used

• These concerns are countered by developments in the program (ex: memos/annotations)

• The coding process is dependent on the analyst knowing and understanding their subject

In Class – Lecture

• 0.5 and 0.1 are the key thresholds/values

o 1-.95 = 1/20 = 0.5 (lie between the inner most bars)

find more resources at oneclass.com

find more resources at oneclass.com

###### You're Reading a Preview

Unlock to view full version

Only page 1 are available for preview. Some parts have been intentionally blurred.

o 1-.99 = 1/100 = .01 (lie outside the outer most bars)

• Which cut-off is better

o 0.5 or .01 – neither is better or worse than the other (read textbook)

• Just as variables have distributions, sample statistics have distribution as well

Calculating P-Values

• Statistics Table – show the values of the distribution functions for different parameters

o How do we know that a z-score of plus or minus 2 translates to p-value of 0.5?

• They used calculus to figure it out how to calculate them by using integrals in university

Statistical Inference

• If e ife soethig it eas that e da a olusio fo it

• Using sample information to estimate the population values (p)

• A tool for drawing conclusions from that are from a random sample

• Probability allows us to take chance variation into account at times

• If e do’t use ado saple, ou olusios a e halleged

• There is absolutely no substitute for collecting good data whatsoever

Types of Distribution

• Sample Distribution - the distribution of a variable using data from our random sample

o The example shown at the beginning of class related to height or grades

• Population Distribution - the distribution of the variable in the population

o We typically do not see this; it is the same as the sample but it is estimated

• Sampling Distribution - the distribution of the mean of the variable

from all possible samples of the same size from the same population

o This is generally the most difficult to understand for most people b/c its abstract

o It is theoretical, you cannot see it but we will look at theories and such to understand

o A sampling distribution is the theoretical distribution of a statistic from all possible

samples of the same size and the same population – different sampling distribution/size

o Its standard deviation gets smaller as the size of the sample tends to get larger

Logic of Statistical Inference Repeated Sampling

• What ould happe if e took a saples?

o Take a large number of samples from the same population

o Calculate the same statistics for each and make a histogram

o Resulting distribution will approximate the sampling distribution

▪ It is a distribution of the means from all possible samples

The Mean of the Distribution of Samples is the Same as the Population Mean?

• Statistical Theory

• Stimulations

o Take a sample (income of 40,000 people)

▪ Population

• Calculate the mean – 1000 random samples, then calculate the mean

of all of the sample means around the mean of the population, then

calculate the mean of the sample means and it will equal the true

population parameter which was previously unknown in the sample

find more resources at oneclass.com

find more resources at oneclass.com

###### You're Reading a Preview

Unlock to view full version