Research Methods and Data Analysis 1 Exam 2

Statistics

Analyzing and making sense of data.

Descriptive statistics

Statistics that organizes, screens for issues, summarizes main features, visualizes with graphs.

Inferential statistics

Statistics that uses data from a simple to infer conclusions about a population.

Population

The total collection of things that we seek information about.

Population parameter

Any summary that describes characteristics of the entire population, this is usually unknown; example might be estimated average weight.

Sample

A representative collection of those things drawn from the population, subset of the population.

Descriptive statistic

Any summary number that describes the sample.

Estimate

Samples differ from each other, this reminds us that we can only ever _____________ population parameters from samples.

Inference

The process of drawing conclusions about population parameters based on a ample taken from the population.

Inferential statistics

There are many cases in which we cannot collect data to test our questions on the full population of interest, because of this we use what?

Shape of the distribution, most common value, the variability of values

To understand the data in a histogram, it is useful to look at what three things?

Inferential statistics

The confidence interval is a measure of what kind of statistics?

Confidence interval (CI)

How far an estimate from the unknown population is from the true mean of that population; a probable range of values for the population estimate; usually 99%, 95%, or 90%.

Confidence interval (CI)

A range of values, that with a known degree of certainty, includes an unknown population characteristics, such as population mean; expresses our "best bet".

Width

The _________ of the confidence interval gives us a probable window for the population parameter.

Narrower

In terms of the width of the confidence intervals, widths that are ______________ are more preferred and are more precise.

By increasing sample size and decrease variation

How can you narrow a confidence interval width?

Sample size and variation (within the population of interest)

What influence confidence interval width?

Narrow

In terms of the confidence interval, populations with low variability lead to __________ confidence intervals, and vice versa.

Population (not sample!)

Confidence interval is only for what kinds of properties?

Fixed

Population value is __________, it is either in a range or not within the range (can't say there is a 90% chance that it is in the range).

Confidence interval

_______________ ________________ says that if repeated samples were taken and a CI calculated, then a blank percentage of the intervals will contain the true mean and then a blank percentage will not contain the true mean; because the true mean is an unknown value we are sure which one they are in.

Confidence intervals

What is one way to bridge descriptive and inferential stats?

Sample

Any smaller collection of actual observations drawn from a population.

Population

Any complete collection of actual observations or potential observations.

Title, abstract, introduction (lit review), methods, results, discussion, references

What are the main sections are an empirical paper?

Title

Section of an empirical paper, very descriptive and often states the main finding.

Abstract

Section of an empirical paper, a brief summary of the study; contains a brief background of the subject, the purpose of the study, basic methods, major findings, and how the findings fit into the field.

Introduction

Section of an empirical paper, presents background information for fellow scientists to understand why the findings are significant; first is the general topic, then it sites past research and identifies gap in the knowledge, then is describes how the study will address that gap and presents hypothesis.

Methods

Section of an empirical paper, should be detailed enough to be replicated, identifies participants, measures, and procedures.

Results

Section of an empirical paper, describes findings and often includes graphs and charts.

Discussion

Section of an empirical paper, summarizes results, how they fit into the filed, and limitations and future directions.

References

Section of an empirical paper, APA style; first author contributed the most, last author listed is the senior author (principle investigator), and the middle authors contributed in some way.

Frequency distribution

A collection of observations produced by sorting observations into classes and showing their frequency (f) of occurrence in each class; helps us detect patterns in the data.

Frequency distribution for ungrouped data

A frequency distribution produced whenever observations are sorted into classes of single values; not always appropriate, mostly used when there are only 20 possible values.

Frequency distribution for grouped data

A frequency distribution produced whenever observations are sorted into classes of more than one value.

Unit of measure

The smallest possible difference between scores; the size of gaps between classes.

Real limits

Located at the midpoint of the gap between adjacent tabled boundaries; that is one half of the unit below the lower table boundary and and one half above the upper table boundary.

Outlier

A very extreme score, potentially impacts the summary of the data.

Relative frequency distribution

A frequency distribution showing the frequency of each class as a fraction fo the total frequency for the entire distribution; allows for relative comparisons; uses proportions or percentages.

Cumulative frequency distributions

A frequency distribution showing the total number of observations in each class and all lower ranked classes; usually converted into percentile ranks.

Percentile ranks

Percentage of scores in the entire distribution with equal or smaller values than that score.

Histogram

A bar type graph for quantitative data; the common boundaries between adjacent bars emphasize the continuity of the data, as with continuous data; consists of a serious of bars whose heights represent frequencies for various classes.

Frequency polygon

A line graph for quantitative data that emphasizes the continuity of continuous variables.

Stem and leaf display

A device for sorting quantitative data on the basis of leading and trailing digits; rarely appear in published reports but useful for organizing data.

Normal distribution

A typical shape of frequency distributions; bell shaped curve.

Bimodal distribution

A typical shape of frequency distributions may reflect coexistence fo two different types of observations in the same distribution.

Positively skewed distribution

A typical shape of frequency distributions; a distribution that includes a few extreme observations in the positive direction (to the right of the majority).

Negatively skewed distribution

A typical shape of frequency distributions; a distribution that includes a few extreme observations in the negative direction (to the left of the majority).

Bar graph

A graph used for qualitative data, gaps between adjacent bars emphasize the discontinuous nature of the data.

Inferential statistics (about population)

Confidence intervals are only used for what kind of statistics?

Hypothesis testing

Testing whether your data support or reject your prediction; involves testing against an alternative possibility using a null hypothesis.

Null effect (effect doesn't exist) or the effect does exist in the population

What are the two possible results of hypothesis testing?

P value

What value is used for hypothesis testing?

Less than 0.05

What value do we want the p-value to be at to conclude that it is statistically significant?

Statistically significant

If the p-value is less than 0.05 we can say that the finding is ________________ ____________.

Frequency distributions

Helps you organize data, see patterns in the data, and detect weird values.

Bins

In frequency distributions there are sometimes to many distinct values for them each to be separate so they must be grouped, to this __________ are used; they are exclusive, equal sized, and exhaustive.

Exclusive

Bins are _____________; each item fits in no more than one bin.

Equal sized

Bins are ________ _________; each bin has the same range.

Exhaustive

Bins are _____________; every item fits into a bin.

Relative frequency distributions

Frequency distribution that uses proportions against total of all frequencies.

Cumulative frequency tables

Frequency distributions that count accumulated scores across bins, useful for count scores below and above a threshold; can be created into percentiles.

Exhaustive, equal sized, exclusive

Bins have what three qualities?

Histogram

Graphical depiction of a frequency table, shows how often values in the data occur.

Scores

What is on the x axis of a histogram?

Frequency

What is on the y axis of a histogram?

Normal distribution

Distribution that is symmetrical with single peak, equal tails, and a bell shape.

Bimodal (or multimodal)

Graphs that have more than one peak.

Skewed distributions

Graphs that have off center peaks, could be positive or negative.

Positive skew

Distribution with tail on the positive side (on the right).

Negative skew

Distribution with a tail on the negative side (on the left).

Outlier

A single point that could change the nature of the distribution and drastically skew results.

Central tendency

Summarizes the middle or most typical values for a variable, has three types; mean, median, and mode.

Mean, median, mode

What are the three types of central tendency?

Level of measurement and shape of distribution

What two factors influence what kind of central tendency to use?

Nominal

Categorical level of measurement.

Ordinal

Rank order level of measurement.

Interval

Level of measurement that has specific increments between variables and no true zero.

Ratio

Level of measurement that has specific increments between variables and has a true zero.

Mode

The most common value of the variable.

Mode

The central tendency of nominal data can only be measured using what?

Discrete

Mode can be used as a central tendency measure for interval and ratio data fi the values are __________; can take a finite number of values.

Median

The middle value in a frequency distribution, the point that falls in the middle of all points when the data is in order; does not use all values in the data (not influenced by other data/outliers).

Ordinal

Using median as a measure of central tendency is best for what level of measurement?

Ordinal, interval, ratio

Median can be used a measure of central tendency for what levels of measurement?

Skewed distributions

What distribution shape is median used for (versus mean)?

Mean

The arithmetic average of all values for a variable, uses all values in the data and is highly influenced by outliers.

Interval and ratio

Mean can be used a measure of central tendency for what levels of measurement?

Balance point

Mean is the ____________ _______ of a distribution; the sum of all scores, expressed as positive and negative deviations from the mean, always equals 0.

Sample mean

The balance point for a sample, found by dividing the sum for the values of all scores in the sample by the number of scores in the sample.

Population mean

The balance point for a population, found by divined the sum for all scores in the population by the number of scores in the population.

Normal

Mean is best when used for data that has what kind of distribution?

Larger

In a positively skewed distribution mean is _____________ than median.

Less

In a negatively skewed distribution mean is ______________ than median.

Variability

Distributions with the same central tendency can have different degrees of ____________.

Between subject, within subject, random measurement error, systematic measurement error

What are the four sources of variability in a distribution?

Participant and measurement

What are the two main sources of variability in a distribution?

Between subject variability

A type of participant variability, differs between people within the group.

Within subject variability

A type of participant variability, differs within given subject; expected for most measures because they are a snapshot of possible scores fro day to day.