So she did the unthinkable. She created a new variable: Grief_Pattern_Categorical (1=Typical, 2=Prolonged, 3=Anticipatory-Inverted). She ran a MANOVA. Then a cluster analysis. Then a two-way mixed ANOVA with time as a within-subjects factor. Each test spat out different results. Each one told a different story. And each time, the ghost of case #089 whispered from the margins, threatening to upend the narrative.
That’s when the first anomaly appeared. trial spss
In the trial SPSS file, she ran a simple linear regression: Grief_Score_Post ~ Grief_Score_Pre + YearsCaregiving . The model output was beautiful. Adjusted R-squared: 0.81. Significance: p < 0.001. But when she scrolled to the casewise diagnostics, row #089 was flagged as an outlier. Studentized residual: -4.2. So she did the unthinkable
“Probably.”
“And your dissertation committee will demand revisions.” Then a cluster analysis
She opened it. Carol’s voice, transcribed verbatim: “People think grief is a straight line. It’s not. It’s a knot. And SPSS can’t untie knots, Doctor. Only hearts can.”
The story began three weeks ago, when her advisor, the gruff and brilliant Dr. Mbeki, had pulled her aside. “Alena, your qualitative data is poetry. But the funding board speaks prose. They want a p-value. They want a significant interaction. Give them a story they can graph.”