01

Veri Dosyasını Yükleyin

Upload your Excel file
veya · or
Tanıtım Videosu
▶ Nasıl Kullanılır?
Bilgi Metni Information Notice · TR
  1. Excel dosyanızın ilk satırında değişkenlerinizi ve maddelerinizi belirtin. Örnek: Tutum (T) → T1, T2, T3. Her madde ayrı sütun olmalıdır.
  2. Veriniz kısaltmalara göre okunacaktır. Eksik veya yanlış maddeleri manuel olarak (+ Ekle) butonundan ekleyebilirsiniz.
  3. Verinizi yükledikten sonra, 2. adımda Güvenilirlik ve Geçerlilik Analizlerini göreceksiniz.
Information Notice EN
  1. Specify your constructs in the first row using: Attitude (AT) → AT1, AT2, AT3.
  2. Variables will be created automatically from these abbreviations. Missing items can be added manually via (+ Add).
  3. After upload, Reliability and Validity Analyses will appear in Section 2.
02

Otomatik Değişkenler

Automatically Created Constructs
Yapılar (Constructs)
Henüz veri yok · No data yet
Kontrol Değişkenleri (Control)
Henüz veri yok · No data yet
03

Veri Önizlemesi

First 100 Rows Preview
0 satır · 0 sütun
Veri yükleyince burada görünecek · Preview will appear here after upload
04

Örneklem Büyüklüğü Hesaplayıcı

Minimum Sample Size Calculator · Hair et al. (2017); Cohen (1992)
33
minimum katılımcı · participants
Cohen/Hair tablosu minimumu: 33
Pratik kural (5 hipotez × 10): 50
Önerilen örneklem: 50 katılımcı
Kaynak · Reference: Hair, J. F., Risher, J. J., Sarstedt, M., & Ringle, C. M. (2019). European Business Review, 31(1), 2–24.  |  Cohen, J. (1992). Psychological Bulletin, 112(1), 155–159.

Measurement Model Analysis

Reliability, Validity & Common Method Bias
2.1

Descriptive Statistics

Skewness & Kurtosis · Normality Check
Item N Min Max Mean SD Skewness Kurtosis Normality
Awaiting analysis
Construct N Min Max Mean SD Skewness Kurtosis Normality
Awaiting analysis
2.2

Factor Loadings

Outer Loadings · Item Reliability
< 0.50 (Weak) 0.50 – 0.70 (Moderate) > 0.70 (Strong)
Construct Item Loading Status
Awaiting analysis
2.3

Reliability & Convergent Validity

Cronbach α · Composite Reliability · AVE
α > 0.7 CR > 0.7 AVE > 0.5
Construct Cronbach α CR AVE
Awaiting analysis
2.4

VIF Values

Variance Inflation Factors · Multicollinearity
≤ 3 (Strong) 3 – 5 (Moderate) 5 – 10 (Weak)
Construct Item VIF
Awaiting analysis
2.5

Cross Loadings

Discriminant Validity Check
Item
Awaiting analysis
2.6

Common Method Bias (CMB)

Full Collinearity VIF & Harman's Single Factor Test
References: Podsakoff et al. (2003); Kock (2015).
Construct-level VIF < 3.3 → no CMB issue · Harman single factor < 50% → CMB not dominant
Full Collinearity VIF (construct average)
Construct Avg VIF
Awaiting analysis
Harman's Single Factor Test (< 50% → OK)
Awaiting analysis

Hypotheses & Model

Structural Model Specification
3.1

New Hypothesis

Define a direct effect path
Info
Each hypothesis represents an arrow in your structural model, tested with β, t, and p-values via bootstrapping.
3.2

Model Paths

Hypotheses List
0 hypotheses
No hypotheses yet
3.3

SEM Model Builder

Auto-Layout Visualization
Constructs are auto-positioned by role (predictor → mediator → outcome). Hypothesis arrows appear automatically as you add them.
Constructs will appear here after data upload
This builder is visualization-only — the analysis uses the hypothesis list above.
3.4

Moderation Effects

Interaction Terms (M × IV → DV)
A moderator changes the strength or direction of the X → Y relationship. The interaction term (M × IV) is automatically added as a hypothesis targeting the DV.

Analysis Results

Structural Model & Discriminant Validity
3.1

Fornell-Larcker

Discriminant Validity · √AVE vs correlations
Diagonal (√AVE) > column correlations → discriminant validity supported.
Construct
Awaiting analysis
3.2

HTMT Matrix

Heterotrait-Monotrait Ratio · Henseler et al. (2015)
HTMT < 0.85 (strict) or < 0.90 (liberal) → discriminant validity supported.
Construct
Awaiting analysis
4

Structural Model — Path Coefficients

Direct Effects · Hypothesis Testing
H Path β (Std) SD t p 95% CI Decision
Hypothesis test results appear here after analysis
4.1

Indirect Effects

Mediation tests
Path β (Std) SD t p Decision
Awaiting analysis
4.2

Total Effects

Direct + Indirect
Total Effect = Direct Effect + Sum of Indirect Effects. If no direct path exists, total = indirect only.
Path Direct β Indirect β Total β Decision
Awaiting analysis
5

R² and Q² Values

Explanatory Power & Predictive Relevance
R²: Explained variance of endogenous constructs (Chin 1998: 0.19 weak · 0.33 moderate · 0.67 strong)
Q²: Predictive relevance via blindfolding (> 0 → predictive relevance supported)
Construct R² adj Interpretation
Awaiting analysis
6

f² Effect Size — Summary

Cohen (1988) thresholds
0.02 (Small) 0.15 (Medium) 0.35 (Large)
Path Effect Size
Awaiting analysis
7

f² Matrix

Effect Size Matrix
From ↓ / To →
Awaiting analysis
8

Model Fit Indices

SmartPLS methodology · Henseler et al. (2014); Dijkstra & Henseler (2015)
Computed from raw data. Green ● Good  |  Orange ● Acceptable  |  Red ● Weak
Index Saturated Model Estimated Model Threshold Status
Awaiting analysis
SRMR: Sun (2005)  ·  d_ULS & d_G: Ringle et al. (2021)  ·  χ²/df: Escobedo Portillo et al. (2016) — between 1 and 3  ·  NFI: Escobedo Portillo et al. (2016) — > 0.90  ·  ρ_A: Dijkstra & Henseler (2015) — ≥ 0.70
9

Path Diagram

Structural Model Visualization
Structural path diagram appears here after analysis
AI Interpretation Contextual insights on your model
Detailed AI-generated interpretation of measurement model, structural model, hypothesis outcomes, R²/Q², effect sizes, and fit indices appears here after analysis.

Confirmatory Factor Analysis (CFA)

Covariance-Based · semopy (lavaan syntax)
What is CFA? Confirmatory Factor Analysis verifies a hypothesized factor structure using covariance analysis. Unlike PLS-SEM (which is variance-based), CFA is covariance-based and typically reports fit indices for the measurement model.

This module uses semopy (Python implementation of lavaan-style syntax). It provides standardized loadings for each item; reliability metrics are computed from the loadings on the frontend.
CFA.0

Prerequisites & Assumptions

Classical CFA assumptions and sample-size rules of thumb
Upload data and define constructs to see the prerequisites check.
Note: These are rules of thumb, not hard requirements. You may run CFA even if some criteria fail, but interpret fit indices and standard errors with caution.
CFA.1

Model Specification

lavaan-style syntax (reflective)
Model specification appears here after running CFA
CFA.2

Standardized Loadings

Std. estimate + flag for items needing review (< 0.70)
< 0.50 (weak) 0.50 – 0.70 (moderate) > 0.70 (strong)
Construct Item Std. Loading Status Flag
Awaiting analysis
CFA.3

Loadings Matrix

Cross-loading style layout
Indicator
Awaiting analysis
CFA.4

Reliability & Convergent Validity

Computed from CFA loadings
α > 0.7 CR > 0.7 AVE > 0.5
Construct Cronbach α CR AVE
Awaiting analysis
CFA.5

Model Fit Indices

Hu & Bentler (1999); Kline (2015)
Fit indices are computed by semopy.calc_stats. Cutoffs: χ²/df < 3 good, < 5 acceptable · CFI & TLI ≥ 0.95 good, ≥ 0.90 acceptable · RMSEA ≤ 0.05 good, ≤ 0.08 acceptable · SRMR ≤ 0.05 good, ≤ 0.08 acceptable.
χ² chi-square
Awaiting analysis
χ² / df < 3 good, < 5 acceptable
Awaiting analysis
CFI ≥ 0.90 acceptable
Awaiting analysis
TLI ≥ 0.90 acceptable
Awaiting analysis
RMSEA ≤ 0.08 acceptable
Awaiting analysis
SRMR ≤ 0.08 acceptable
Awaiting analysis