PSYC TEST CH 6-10

How do we calculate a related / paired samples t test

1. State hypothesis 2. Set decision criterion ( typically for behavioral research we set alpha level at .05). 3. Collect Sample Data Exm. The study collected sample from 50 psychics, they all participated in prediction tasks in both rooms 4. Calculate statistics- estimated standard error, calculated t-value, degrees of freedom and the critical t value is t. 5. Make a decision

Stating the hypothesis

1. State hypothesis Alternative hyp.: in the population the mean difference score should be greater than 0, note that this is a one-tailed test Null hyp: in he population, the mean difference score should be less than or = to 0 We compare the sample mean to 0 because if there is no effect of condition orb dependent variable, then we expect no difference between the scores from these conditions. U^1. =u^2

Creating Bar Graphs

1. Transform raw data into descriptive statistics. 2. Highlight the cells in the window. 3. Choose the bar option (under CHART function from the insert menu.

Creating Line Graphs

1. Transform the raw data into descriptive statistics 2.Highlight the cells in the window 3. Choose the X Y SCATTER option ( under CHART function from the INSERT menu). 4. To connect them with the lines: A. Click on Change chart type. B. Choose Scatter with Lines (under day scatter graphs options Click the button to switch ROW/COLUMN TO PUT THE CONTINUIOS GROUPS ON TH X-AXIS AND CONTINUOUS OUTCOME ON THE Y-AXIS.

Sections of Research Articles

Abstract- summary of the study Introduction- a description of the topics of the study Methods: details of participants, methods, and procedures Results: a description of analysis, descriptive and inferential statistics. Discussion: a review of the results with possible future directions and general conclusions.

Example of variable

Age, place in a race, or gender because it takes on 2 different values

Categorical

Allow for classification based on characteristic. You cannot add or subtract or divide exp: gender variables

Difference between bar graphs, line graphs and pie chart

Bar graphs organize data and show results of each different factor. Line graphs Pie charts give a visual of the ratios of my data.

How to choose best way to present data

Based on rather variables are categorical or Continuous

Graphs and type of data

Categorical variables and Continuous variables Categorical variables: responses are categorical as in nominal and ordinal scale of measurement that cant be divided into smaller units Usually presented in pie charts or bar graphs Continuous variables- responses are numerical scores that can be subdivided into smaller units. Usually presented in line graphs

Descriptive Statistics in graphs

Describes data

The ways to present data from a study

Graphs and tables

Different ways to present data

Graphs and tables Graphs show aspects of data more visually and clearly Tables presents more precise values that graphs typically don't.

Graphs and Tables in APA Styles

Graphs are referred to as Figures, with each fixture on its own page with a title. A description of the figure or table is also provided in the results section where the author references these figures Tables cannot be pasted from SPSS or Excel output If you create a Table in word, you will have to format the table to remove the vertical lines.

Pie Charts

Graphs of categorical variables where proportions are represented as portion of the "pie" It is not used as often as bar and line graphs in reporting data They are useful in presenting percentage responses data for simple surveys from different categories of responses or respondents.

Bar graphs

Graphs of data for categorical variables where the bar height represents the size of the value Sometimes bar graphs include error bars which show the 95% confidence intervals for each group.

Line Graphs

Graphs of data for continuous variables where each value is graphed as a point and the points are connected to show differences between scores Graph the values for each group or condition as points on the y-axis centered above the appropriate value of the continuous variable on the x-axis.

Scatter plots

Graphs showing the relationship between two dependent variables for a group of individuals

How can we compare data from two different situations

Samples with related/ paired data

Difference scores

Simply subtract the score for each condition from the other condition Calculate the mean of the difference scores, by adding up all the difference scores and dividing by the total number of subjects/ pair

APA STYLE AND ITS IMPORTANCE

Structure followed by most psychological research articles Some sections mo=Ishtar be omitted depending on the length of the article.

What is the difference between aa within-subjects design and a matched pairs design?

The data are collected for single sample in two environments / conditions One common within-subjects design is pretext- posttest design Pretext/ posttest design is a type of research design (often a quasi-experiment) where behavior is measured both before and after a treatment or condition is implemented.

Related / paired samples t test

We compare across scores that are related in some way The steps to the hypothesis test is very similar to the one-sample t-test, except we will use the difference score of the pairs of scores rather than the raw scores.

Matched Design

a between-subjects experiment that involves sets of participants matched on a specific characteristic with each member of the set randomly assigned to a different level of the independent variable

Descriptive Statistics in Tables

more likely to see tables that present descriptive statistics from data sets It is common to see the table of central tendency measures by level of the variables present in the design of study. Descriptive statistics should be presented in the format that best displays th test of the hypothesis or conclusions made from the data. Descriptive statistics should not be repeated as both a graph and a table. If a graph is the best way to display data, do not include table. Tables and graphs are meant to help the audience better understand a summary of the data

Inferential statistics

Last step of putting descriptive statistics in tables Presented into text to provide the hypothesis test

When creating pie charts in excel

Transform the raw data into descriptive statistics (frequency counts). Highlight the cells in the window Choose the PIE OPTION (UNDER CHART FUNCTION from the insert menu) OPTION: Include the legend of the letter grades by choosing the LEGEND option under Chart Design

Test assumption 1

The population of difference scores must be a normal distribution This assumption is necessary in order to ensure that the values in the t table are accurate; however, when a sample size n is larger than 30 and the sample was selected at random from the population, violating this assumption does not change the critical t values enough to change the outcome of participants must be independent.

Test assumption 2

The scores from different participants or pairs of participants must be independent. This means that across the subjects or across the Pair s of subjects for matched designs the scores cannot depend on the scores of the other participants or pairs.

CREATING GRPHS IN SPSS

This is easier than excel