SPSS Data Analysis Quiz

Podcast

Play an AI-generated podcast conversation about this lesson

Questions and Answers

What does the 'percent' column in the output table of SPSS frequencies represent?

The cumulative total of participants across all categories

The total number of participants in the study

The proportion of participants in each category relative to the total sample (correct)

The percentage of valid data points collected only

What should be done with missing data in SPSS?

Always remove missing data from the analysis

Create impossible values and instruct SPSS to ignore them (correct)

Leave missing data as is for accurate representation

Substitute missing values with the mean of other data

Which of the following analyses is NOT one of the three main types mentioned for running simple analyses in SPSS?

Normal distribution

Frequencies

Inferential statistics (correct)

Descriptive statistics

What feature of SPSS allows you to change the physical width of the column in data view?

Columns Signup and view all the answers

What does the valid percent indicate in the output of SPSS frequencies?

Percentage without accounting for missing datapoints Signup and view all the answers

What does a smaller p value indicate in statistical analysis?

Lower likelihood of type I error Signup and view all the answers

What is the alpha level set at in most psychological studies?

0.05 Signup and view all the answers

Which assumption is not typically associated with parametric analysis?

Ordinal data collection Signup and view all the answers

How does sample size affect the significance of a difference between conditions?

More participants increase the likelihood of finding significant results Signup and view all the answers

What does violating the assumption of normal distribution imply for a dataset?

Data might be skewed leading to potential bias Signup and view all the answers

What characterizes a leptokurtic distribution?

Majority of scores are concentrated in the middle Signup and view all the answers

Which of the following is true about the homogeneity of variance assumption?

It posits variance should be similar across all conditions Signup and view all the answers

What does a positively skewed distribution indicate?

The peak of the frequency falls at the higher end Signup and view all the answers

What effect does committing a type I error have in hypothesis testing?

Rejecting a true null hypothesis Signup and view all the answers

Which type of data should be collected for robust results in parametric analysis?

Interval or ratio data Signup and view all the answers

What is recommended regarding research questions in content analysis?

They should have a specific predefined question. Signup and view all the answers

What influences the determination of the material for analysis in research?

The specifics of the research question. Signup and view all the answers

Which of the following best describes the recording unit in content analysis?

An individual word or specific detail. Signup and view all the answers

What should be ensured for categories during content analysis?

Categories should be clear and distinct. Signup and view all the answers

How can categories in content analysis be formulated?

Through either top-down or bottom-up approaches. Signup and view all the answers

What can happen as a result of poorly coded categories in content analysis?

Lower inter-rater reliability. Signup and view all the answers

What is the first step in deciding the recording unit for content analysis?

Define the unit of measurement. Signup and view all the answers

What does the meticulous process of identifying how to code and categorize content involve?

Determining the recording unit. Signup and view all the answers

What should be reported alongside the t statistic when writing up results?

Degrees of freedom Signup and view all the answers

When is it appropriate to use the 'equal variances not assumed' row in a t test?

When Levene's test is significant Signup and view all the answers

What does a negative t value indicate?

Higher scores in the second condition Signup and view all the answers

Which effect size is commonly used in conjunction with a t test?

Cohen's d Signup and view all the answers

What output does the Mann-Whitney U test provide?

Ranks and test statistics Signup and view all the answers

What is the primary requirement for using a repeated measures t test?

Only two conditions Signup and view all the answers

If a p value comes back as .000, how should it be reported?

p < .001 Signup and view all the answers

Which software option is used to run a Wilcoxon test?

Analyse → Nonparametric Tests → Legacy Dialogues → 2 Related Samples Signup and view all the answers

What does 'sphericity' refer to in the context of repeated measures t tests?

Equal variances among conditions Signup and view all the answers

What is the role of Levene's test in t tests?

To assess equality of variances Signup and view all the answers

Which of the following is a requirement for using the Mann-Whitney U test?

Ordinal or non-normally distributed data Signup and view all the answers

What statistical output is necessary for understanding the repeated measures t test results?

Paired Samples Correlations and Paired Samples Stats Signup and view all the answers

What happens to the degrees of freedom when the assumption of equal variances is violated?

They are adjusted to be smaller Signup and view all the answers

What is a key characteristic of thematic analysis?

It acknowledges and interprets themes within qualitative data. Signup and view all the answers

What role do memos play in grounded theory analysis?

They assist in coding and capturing analytical reflections. Signup and view all the answers

What does the open coding process involve?

Initial breakdown of transcripts to create concepts. Signup and view all the answers

Which type of analysis transforms qualitative data into quantitative summaries?

Content analysis. Signup and view all the answers

Why is interrater reliability important in qualitative research?

To minimize human bias and ensure consistent coding. Signup and view all the answers

What is a major concern regarding the use of software packages in qualitative analysis?

They may overlook the context of qualitative data. Signup and view all the answers

What is the essence of a thematic map?

It visually represents candidate themes and their relationships. Signup and view all the answers

What does it mean to conduct semantic analysis in research?

It simplifies the approach and is suitable for beginners. Signup and view all the answers

What is a defining feature of grounded theory?

It focuses on generating new theories from data. Signup and view all the answers

In what way should the themes developed in thematic analysis be evaluated?

They must make sense within the entire dataset's context. Signup and view all the answers

What is an essential consideration when defining sample size in qualitative research?

It can be related to the volume of data generated by participants. Signup and view all the answers

What is a common challenge when using CAQDAS for qualitative analysis?

It sometimes overlooks the qualitative context of the data. Signup and view all the answers

What is the final phase in thematic analysis focused on?

Writing a clear and valid representation of the dataset. Signup and view all the answers

Study Notes

Significance and P-Value

Significance in statistical analysis is determined by the p-value.
The p-value represents the probability of obtaining the observed results if there is no real effect.
Smaller p-values indicate stronger evidence against the null hypothesis and suggest statistical significance.
The alpha level (typically set at 0.05) represents the threshold for determining significance.
A p-value less than the alpha level is considered statistically significant, meaning there is a low probability of observing the results by chance.

Degrees of Freedom

Degrees of freedom (df) are influenced by sample size, affecting significance.
Larger sample sizes generally lead to higher degrees of freedom, making it easier to find statistically significant results.
A difference of the same size between conditions can result in different significance findings due to differing sample sizes.

Parametric Analysis (PA) Assumptions

PA is considered the gold standard for inferential statistics.
PA depends on specific assumptions about the data and methodology.
Four key assumptions vital to PA: independence of observations, interval or ratio-level data, normal distribution, and homogeneity of variance.

Independence of Observations

Ensure that each participant's data is independent of others, meaning one participant's data does not influence another's.
This assumption is usually easy to maintain during data collection.

Interval or Ratio-Level Data

PA requires data measured at the interval or ratio level.
Interval and ratio scales provide more precise information than nominal or ordinal scales.

Normal Distribution

Data should ideally follow a normal distribution, characterized by a bell-shaped curve.
Most participants should obtain scores around the middle of the distribution, with scores decreasing symmetrically on either side.
Deviations from normality can be caused by skewness or kurtosis, requiring specific statistical adjustments.

Homogeneity of Variance

Assumes that the variance in the dataset is similar across all conditions.
The magnitude of the variance itself is not critical, but it should be consistent across all conditions.

Missing Data Handling in SPSS

Missing data can be addressed using SPSS by creating impossible values and instructing the software to ignore them.

Variable Definition in SPSS Variable View

It is recommended to define variables in SPSS Variable View by working systematically through each variable and its options.

Running Simple Analyses in SPSS

Three main analyses: frequencies, descriptive statistics, and tests for normal distribution.
These analyses share similar dialogue boxes in SPSS.

Frequencies Analysis in SPSS

Helps determine the number of participants in specific categories or groups.
SPSS output provides frequency counts, percentages (including valid and cumulative percentages).
Typically, report both frequency counts and percentages.

Descriptive Statistics in SPSS

Provides a summary of continuous data or compares groups on continuous variables.
SPSS outputs descriptive statistics with separate rows for "equal variances assumed" and "equal variances not assumed," representing different degrees of freedom.
The Levene's test determines which row to use: significant Levene's test indicates unequal variances, so use the row "equal variances not assumed."
Output includes essential information for reporting t-tests.

Reporting T-Test Results

The t-statistic, degrees of freedom (df), and p-value are reported.
Follow the format: "t (df) = XX.XX, p <.XXX".
The t-statistic (XX.XX) indicates the magnitude and direction of the difference between groups.
The p-value (.XXX) indicates the probability of observing the results by chance.

Effect Size in T-Tests

Reporting effect size alongside the p-value helps interpret the practical significance of the findings.
Cohen's d is commonly used for t-tests.
Effect size categories:
- Small effect: ≥ 0.2
- Medium effect: ≥ 0.5
- Large effect: ≥ 0.8

Non-Parametric Equivalent for Independent Samples T-Test: Mann-Whitney U

Used when assumptions are violated (e.g., non-normal DV).
Converts data into rank scores, enabling testing without distributional assumptions.
SPSS output provides the U statistic and p-value, similar to the t-statistic.
Do not report df or N.

Repeated Measures T-Test

Used when comparing data from the same participants across different conditions.
Data is structured differently compared to independent samples t-tests.
Sphericity assumption is tested instead of homogeneity of variance (cannot be computed for two conditions).

Running a Repeated Measures T-Test in SPSS

Select both variables and analyze the data using the Paired-Samples T Test function in SPSS.

Interpreting Repeated Measures T-Test Output

Focus on the "Paired Samples Stats" table (provides descriptive statistics) and the "Paired Samples Test" table (contains t-value, df, and p-value).
If the p-value is < .001, report it as "p <.001."

Reporting Repeated Measures T-Test Results

Follow the format: "t (df) = XX.XX, p <.XXX."

Effect Size for Repeated Measures T-Test

SPSS does not provide Cohen's d for repeated measures.
Use the Cohen's d table provided in the notes for interpretation.

Non-Parametric Equivalent for Repeated Measures T-Test: Wilcoxon Signed-Rank Test

Used when DV is ordinal or conditions have non-normal distributions.
SPSS output provides ranks (descriptive stats) and test statistics (z score, p-value).
Do not report df or N.

One-Sample T-Test

Compares one sample to a known or theoretical reference value.
Used when you want to see if the sample differs from a specific value.
SPSS output provides t-value, df, and p-value.

Content Analysis

A method of systematically analyzing text, audio, or video data.
Involves coding and categorizing the content based on pre-defined or emergent categories.
Used to understand the content of communications, identify themes, or explore patterns.

Stages of Content Analysis

Define Research Question and Identify Material:
- Clearly define the research question and identify the appropriate body of material for analysis.
Decide on Recording Unit and Categories:
- Define the unit of measurement (e.g., word, phrase, sentence) and develop clear categories for coding the data.
Code and Analyze the Data:
- Systematically code the data using the defined units and categories.
- Analyze the coded data to identify patterns and themes.

Inter-Rater Reliability

Measure of agreement between coders regarding the categorization of content.
Important for ensuring consistency and reliability in content analysis.

Main Analysis of Qualitative Data

Following data collection, the main analysis of qualitative data begins.
The analysis should be conducted by multiple coders to ensure interrater reliability.
The researcher should use the complete corpus of materials predefined at the start of the study.
Findings should be explored in relation to additional variables that may have formed secondary hypotheses.
Statistical reporting should follow APA format guidelines.

Software Analysis

Software analysis uses a program to analyze data.
CAQDAS is a software program that can analyze qualitative data and transform it into a quantitative summary.
CAQDAS can be used for coding and categorizing, but it is important to consider potential issues of context when applying it.

Thematic Analysis

Thematic analysis is a foundational qualitative technique used across many disciplines.
It involves identifying and analyzing themes within qualitative data, and is frequently used with a variety of data collection methods.
The thematic analysis may be deductive or inductive.
The researchers must interpret the themes.
Semantic analysis is a common and relatively simple approach to thematic analysis.
It is a good starting point for those new to qualitative analysis.

Types of Research Questions

Thematic analysis is suitable for research questions that seek to identify themes within a particular group or population.

Sample Size in Thematic Analysis

Thematic analysis sample size is determined by the overall size of the corpus of data, not just the number of participants.
It is not always necessary to have a large number of participants.
A smaller number of participants with a substantial amount of data can be sufficient.
It is less common to define a sample size in advance.

Phases of Thematic Analysis

Once the question and data collection method are decided, there are six phases in thematic analysis:

Familiarization with the Dataset

It is crucial to fully read the data.
This may involve transcribing verbal data to written form.

Coding Individual Data Items

Codes should describe the data relevant to the research question.
Codes can be manifest (explicit) or latent (implicit).
Codes may be theory-driven or data-driven.
Codes can be interpretative.

Searching for Themes

This involves comparing and collating codes to identify potential themes.
Various techniques can be employed.
The aim is to produce a number of candidate themes.
A thematic map can be created to visually represent themes and the relationship between codes.

Reviewing Themes

This involves evaluating themes to ensure they make sense in relation to the data set.
Themes may be broken down, combined, or discarded.
The researcher should check each code and its associated data extracts to ensure they fit the theme.
The researcher should review the data set to confirm that the thematic map makes sense in context.

Naming and Defining Themes

The researcher should identify the essence of each theme.
Subthemes should be clearly defined.
Theme names should encapsulate the overarching meaning, not just a one-word name.

Writing Up

The write-up should provide a clear and valid representation of the data set.
The write-up should include core aspects of a psychology report.
Given the flexibility of semantic analysis, the write-up should clearly outline the epistemological background that informed the study.
Different analytic methods may have varied write-up procedures.

Thematic Analysis Write-up

The write-up should include an introduction to analysis, an analytical overview, and a presentation of analysis.

Introduction to Analysis

Reiterate the type of analysis used.
Provide a description of specific transcription symbols.

Analytical Overview

Outline the themes and subthemes.
Avoid using the term "emergent theme," as it implies the themes were already present.
State the number of themes and list them in a logical manner.
A thematic map can be used to depict relationships between themes and subthemes.

Presentation of Analysis

Provide detailed information about each theme.
Outline the theme and reiterate subthemes.
Provide examples with quotes.
Indent and separate quotes from the main text.
Interpret each quote and connect it to the literature (either after the quote or in the discussion section).

Key Considerations: Ensuring Themes are Grounded

Qualitative data is often transformed into quantitative data through systematic and exhaustive definition of units and categories.
Content analysis strives for interrater reliability by collecting and analyzing frequency data.
Thematic analysis is founded in qualitative research methods, but acknowledges the importance of context.
When conducting semantic analysis, the researcher must consider the social constructionist perspective.

Ch24: Grounded Theory

Grounded theory is a qualitative technique for developing theory from data.
It is an overarching approach to study design that aims to generate new theory.
It is inductive and aims to develop substantive theory about a phenomenon in a specific context.
There are many variations within this approach.

Types of Research Questions for Using Grounded Theory

Grounded theory is especially suited to research questions about topics that are new or understudied.
The focus of the study should be on a psychosocial phenomenon.
Grounded theory aims to explain a phenomenon rather than confirm a hypothesis.

Key Features and Terms in Grounded Theory

Grounded theory involves:
- Concurrent data collection and analysis
- Focus on psychosocial processes
- Codes and categories developed from the data
- Memos to aid analysis
- Inductive production of abstract categories
- Theoretical sampling
- Incorporation of categories into a theoretical framework

Continuous Process of Collection and Analysis

Grounded theory involves continuous collection and analysis of data.

Sampling and Collecting Data

Grounded theory utilizes purposive sampling to find information-rich cases of the phenomenon.
The researcher makes subjective decisions on who to include in the study.
Data is usually collected through interviews or focus groups.
Open-ended or semi-structured interviews are common.
Only relevant bits of data should be transcribed.

Producing Memos

The researcher should create memos throughout the research process to record thoughts on potential codes and interpretations.
Memos are a central part of the analytical process.
Memos help analyze data and develop a theory.

Analyzing the Data: Different Coding Types and Category Development

Grounded theory involves coding data at three levels: open coding, axial coding, and selective coding.
Codes are grouped together to represent concepts.
Concepts are further grouped to form categories.
Theory is developed by describing the relationships between categories.

Open Coding

This initial stage involves breaking down the transcripts to generate initial codes and concepts.
Codes can be words, sentences, or paragraphs that describe a particular feature of the phenomenon.
The code must be meaningful in the context of the data.
Open coding conceptualizes data at a descriptive level.
Initial open codes should attempt to analyze the data.
Theoretical sensitivity is essential.
Two types:
- "In vivo" codes: These use the participant's specific words.
- "Constructed" codes: These are more conceptual.

Grouping Open Codes as Concepts

Open codes are grouped together to identify concepts.
Concepts represent a higher level of abstraction.
All relevant codes are summarized as a single concept.
Groups of concepts form categories.

Constant Comparison

Constant comparison is an essential part of grounded theory.
The researcher compares and contrasts codes and categories with the data to identify similarities and differences.
The flip-flop technique involves comparing a concept with its hypothetical opposite.
Constant comparison leads to revisions and relabelings of codes and categories.
Theoretical saturation occurs when new codes consistently fit into existing categories and no new categories are formed.
The researcher should continue constant comparison until theoretical saturation.

Abstract Definitions

Formal definitions of each category are developed at an abstract level.
These definitions should include:
- Characteristics of each category
- Defining psychological constructs
- When things occur, how they manifest, and the consequences of their occurrence

Theoretical Sampling at Every Stage

Theoretical sampling involves collecting data to generate theory.
The researcher collects data, codes and analyzes, and then makes decisions about what data to collect next.
Theoretical sampling should begin early in the research process.
Identify a purposive sample and then seek participants that confirm the developing categories.

Axial Coding

Axial coding involves identifying relationships between categories by exploring connections between codes, linking codes to create a more comprehensive understanding of the phenomenon.
A coding paradigm is used to stimulate hypotheses and questions.
Axial coding reduces the number of categories.

Selective Coding and Identifying a "Core Category"

Selective coding explores relationships at the conceptual, property, and dimensional level.
The researcher identifies the core category that brings together all the categories.
The core category is typically abstract but grounded in the data.

Theoretical Integration

Theoretical integration involves producing a storyline that represents the relationships between core categories and other categories.

Grounding the Theory and Filling in the Gaps

The researcher should validate the theory against the data.
Each statement of the theory should be reviewed for support.
The write-up stage follows a standardized format.
The results section is typically the most substantial part of the write-up.
Abbreviated grounded theory may only involve coding and constant comparative analysis.

Ch25: Interpretative Phenomenological Analysis (IPA)

IPA examines the core structures of individual experiences and how individuals make sense of them.
It places emphasis on the meanings people hold about their experiences.
It is rooted in hermeneutics and phenomenology.

Interpretation and Hermeneutics

The aim of hermeneutics is to understand lived experience and to ask questions about interpretation.
The focus is on the perception of the experience, not truth checking or validation.
The analyst may offer interpretations that the participant may not be able to.
Interpretation is a natural human activity.

Descriptive vs. Interpretive Phenomenology

There are two types of phenomenological approaches:
- Interpretive Phenomenology: Focuses on the sense-making activity of the participant.
- Descriptive Phenomenology: Focuses on exploring the essence of the experience and its structure.

Descriptive Phenomenology and Husserl

Developed from Husserl's perspective on the scientific study of individual experiences.
Argues subjective experiences are worthy of attention.
AIMS to describe lived experiences.

Interpretive Phenomenology and Heidegger

Argues that we should go beyond description and search for meaning embedded in common life practices.
May be inductive or deductive in its approach to using theory in the research.
The final analysis is derived from both participant and research interpretations (double hermeneutic cycle).

Commitment to Idiography

IPA is committed to idiography in two ways:
- Detailed, in-depth analysis
- Understanding how experiential phenomena are understood.

Research Questions and Data Types

Research questions should explore an experiential event.
Questions are often open-ended to allow for detailed exploration.
IPA is suited to research exploring unique experiences.

Data Collection Methods

Typically semi-structured interviews are employed.
Other methods include:
- Photo elicitation
- Focus groups
- Secondary sources

Sample Size

Sample size is small to ensure each participant has a chance to offer insight.
Participants should be homogenous.

Interview Schedule + Conduction

The interviewer should try to understand the participant's perspective and their experience.
The interview schedule should be flexible.
The interviewer should foster an empathetic and sensitive manner.
Begin by asking participants to describe the experience and then follow up with exploratory questions.
Avoid interrupting.
Interviewing is often used with sensitive topics.

Stages of Analysis

The analysis should be consistent with the epistemological and philosophical roots of IPA.
Adaptations and deviations are acceptable as long as the analysis stays consistent.

Transcribing Data

Transcribe data as soon as possible.
Use verbatim transcription.
View transcription as the first stage of analysis.

Stage 1: Reading and Re-Reading

Read the transcript slowly and consider the words and phrases used.
Start making initial interpretations.

Stage 2: Initial Noting

Examining the transcript to develop initial codes.
Free text analysis is used.
Commentary should be honest and instinctive.
There should be an evident phenomenological focus.

Stage 3: Developing Initial Themes

Themes are formed from clusters of discrete chunks of the transcript and associated commentary.
Create a table of initial themes.

Stage 4: Searching for Connections Across Initial Themes

Print out a list of themes, cut them up, and sort them into piles.

Stage 5: Moving to the Next Case

Repeat stages 1-4 on each case.

Stage 6: Looking for Patterns Across Cases

Identify patterns across cases from the tables of initial themes.
Larger groups of themes are grouped together to form a master theme.

Reflexivity

The researcher is also an interpretive being.
Involve the reader in the researcher's reflexive process.

Ch26: Discourse Analysis

Discourse Analysis is an interdisciplinary approach that analyzes how people use language to construct meaning.
It is embedded in a social constructionist approach.

What is Discursive Psychology

One specific approach within the umbrella of discourse analysis.

Action Orientation

Discursive psychology examines how people use language to achieve a particular action.
Rather than inferring what people think, the focus is on what people say and what language is doing within an interaction.
Context is important for understanding language.

Stakes

The way people talk varies depending on context.
People are aware of their "stakes," which include desires, beliefs, and loyalties.
Discursive psychologists are interested in how people manage their stakes within interactions.

Interpretive Repertoires

Refers to a coherent way of speaking about something.
These repertoires may rely on clichés.
There may be tension between opposing viewpoints.
Discursive psychology examines how people negotiate dilemmas in everyday interactions.

When to Use Discursive Psychology

Use discursive psychology to explore how people construct their experiences.
Focus on what people are doing with their language.
Discursive psychology researchers narrow their focus to specific aspects of language.

Studying That Suits You

Use AI to generate personalized quizzes and flashcards to suit your learning preferences.

Description

Test your knowledge on key features and analyses in SPSS with this quiz. Questions cover topics such as the interpretation of frequency tables, handling missing data, and various analysis types. Perfect for students and professionals looking to enhance their SPSS skills.