SPSS Data Management and Analysis Training Course

Participants can define variables and assess whether a dataset supports valid analysis.

• Cases, variables and records within rectangular SPSS data files
• Nominal, ordinal and scale measurement levels for procedure selection
• Variable names, labels and value labels following naming conventions
• Missing-value codes distinguished from valid zero and blank values
• Date, string and numeric formats matched to source data
• Codebooks documenting source fields, definitions, ownership and permitted values

Participants can import external files and combine datasets using documented identifiers.

• Excel and CSV import settings for headers, types and encoding
• Import previews used to identify malformed dates and numeric fields
• Unique identifiers checked before merging customer, survey or programme files
• ADD FILES combines records with matching variable structures
• MATCH FILES joins attributes using sorted identifiers and merge keys
• Post-merge frequency checks reveal unmatched records and duplicate identifiers

Participants can identify data-quality defects and specify defensible correction rules.

• Frequencies identify impossible codes, blanks and unexpected category values
• Descriptives reveal extreme minimums, maximums, means and standard deviations
• SORT CASES locates duplicate identifiers and suspicious repeated records
• Range checks compare entered values against business rules
• Cross-field checks test logical consistency between related variables
• Data-quality logs record issue, decision, rationale and affected records

Participants can create analysis-ready variables while retaining clear transformation logic.

• RECODE converts raw categories into documented analytical groupings
• COMPUTE creates ratios, scores, flags and derived indicators
• IF conditions apply targeted transformations without changing unrelated cases
• DO IF structures handle multi-branch business rules transparently
• Automatic recoding converts string categories into numeric analysis codes
• Variable labels preserve definitions for every derived analytical measure

Participants can write and run syntax that documents repeatable SPSS procedures.

• Syntax Editor commands, comments, line endings and command execution
• Paste function captures menu selections as editable syntax
• GET FILE and SAVE OUTFILE establish reproducible data paths
• Variable labels and value labels assigned through syntax
• Temporary transformations controlled using TEMPORARY and SELECT IF
• Annotated syntax sections support peer review and future updates

Participants can control analytical populations and apply weights with appropriate caution.

• SELECT IF defines analysis populations using documented inclusion criteria
• FILTER creates reversible subsets while retaining excluded source records
• SPLIT FILE produces separate output for regions, groups or periods
• WEIGHT BY applies survey or population weights correctly
• Unweighted and weighted results compared before reporting differences
• Selection settings checked to prevent unintended restricted analyses

Participants can assess missingness and outliers before selecting treatment approaches.

• Missing-value patterns examined with frequencies, tables and visual inspection
• Listwise and pairwise deletion compared for analytical consequences
• System-missing and user-missing values handled in SPSS procedures
• Boxplots identify potential univariate outliers requiring investigation
• Z-scores flag unusually distant observations under stated assumptions
• Decision logs distinguish correction, exclusion, retention and imputation choices

Participants can summarise variables and assess distributions for subsequent analysis.

• Frequencies report counts, percentages, modes and category distributions
• Descriptives calculate mean, standard deviation, minimum and maximum values
• Explore produces medians, percentiles, confidence intervals and normality plots
• Histograms assess shape, skewness, modality and data-entry anomalies
• Boxplots compare spread and central tendency between groups
• Distribution summaries matched to nominal, ordinal and scale variables

Participants can produce clear SPSS tables and charts suited to decision questions.

• Chart Builder creates bar, line, pie, histogram and boxplot charts
• Chart titles state measure, population, period and comparison clearly
• Axes, labels and category order revised for accurate interpretation
• Error bars display uncertainty around group means when appropriate
• Custom Tables produce controlled summaries for management reporting
• Output tables edited to remove clutter and retain essential evidence

Participants can analyse relationships between categorical variables and interpret association measures.

• CROSSTABS compares category distributions across defined population groups
• Row, column and total percentages selected for the question
• Chi-square tests assess association between categorical variables
• Expected cell counts checked before accepting chi-square results
• Phi and Cramer's V quantify categorical association strength
• Standardised residuals identify cells contributing most to associations

Participants can interpret statistical uncertainty and distinguish practical from statistical findings.

• Sampling distributions explain uncertainty in estimates from observed samples
• Confidence intervals communicate plausible ranges around estimated population values
• Null and alternative hypotheses written before selecting statistical tests
• P-values interpreted alongside effect sizes and confidence intervals
• Type I and Type II errors considered in decisions
• Statistical significance distinguished from material operational importance

Participants can select and interpret two-group comparisons using parametric and non-parametric methods.

• Independent-samples t-tests compare means for separate participant groups
• Paired-samples t-tests assess change between matched measurements
• Levene's test checks equality of variance assumptions
• Cohen's d expresses standardised difference between group means
• Mann-Whitney U tests compare independent ordinal or non-normal data
• Wilcoxon signed-rank tests assess paired non-parametric differences

Participants can analyse differences across three or more groups and identify meaningful contrasts.

• One-way ANOVA compares means across three or more groups
• Homogeneity of variance assessed using Levene's test output
• Post-hoc tests identify group differences after significant ANOVA
• Eta squared reports variance explained by group membership
• Kruskal-Wallis tests compare multiple independent non-normal groups
• Effect plots communicate group means and confidence intervals

Participants can quantify and report relationships between two measured variables.

• Scatterplots inspect form, direction, spread and unusual observations
• Pearson correlation measures linear association between scale variables
• Spearman correlation assesses ranked or non-normal variable relationships
• Correlation coefficients interpreted by direction, magnitude and uncertainty
• Correlation does not establish causation or control confounding factors
• Bivariate findings written with variables, sample size and significance

Participants can build and interpret a multiple linear regression model for a continuous outcome.

• Regression questions identify continuous outcomes and candidate predictor variables
• ENTER method estimates specified predictors within a theory-led model
• Model Summary reports R squared and adjusted R squared
• ANOVA table tests overall regression model statistical significance
• Coefficients table interprets unstandardised, standardised and significance estimates
• Prediction equations translate regression output into operational scenarios

Participants can evaluate linear regression assumptions and report model limitations.

• Residual plots assess linearity, homoscedasticity and unusual fitted patterns
• Normal probability plots inspect residual distribution assumptions
• Variance inflation factors identify problematic multicollinearity among predictors
• Cook's distance flags influential cases requiring case-level review
• Standardised residuals identify observations poorly predicted by the model
• Model limitations documented without overstating predictive or causal claims

Participants can assess multi-item measures and construct transparent composite scores.

• Cronbach's alpha assesses internal consistency for multi-item scales
• Item-total correlations identify items weakly aligned with a scale
• Reverse-coded items corrected before calculating composite measures
• MEAN and SUM functions build documented scale scores
• Minimum answered-item rules handle partially completed questionnaires consistently
• Reliability results interpreted alongside construct definition and item content

Participants can analyse survey and indicator data for programme, service and operational questions.

• Survey questions mapped to variables, indicators and analytical populations
• Response rates calculated using stated denominator and eligibility rules
• Disaggregation compares outcomes by location, demographic group or service channel
• Baseline and follow-up measures aligned before change analysis
• Indicator definitions protect consistent reporting across reporting periods
• Findings distinguish observed patterns from unsupported attribution claims

Participants can communicate findings, limitations and recommendations without overstating evidence.

• Analysis plans link decisions, questions, measures, methods and outputs
• Executive findings state result, magnitude, implication and recommended action
• Technical appendices document data preparation, tests and assumptions
• Tables and charts selected according to intended audience needs
• Limitations cover data quality, sample scope, bias and uncertainty
• SPSS output exported to Word, Excel and presentation formats

Participants can build, review and present a complete SPSS analysis from raw data to actionable recommendations.

• Business question translated into outcomes, predictors, groups and analytical hypotheses
• Raw dataset imported, screened, cleaned and documented using SPSS syntax
• Derived variables created with transparent recoding and calculation rules
• Descriptive, comparative and relationship analyses selected from evidence requirements
• Diagnostic checks and limitations recorded alongside statistical findings
• Management-ready charts, tables, conclusions and recommendations prepared for presentation