fix sample size

This commit is contained in:
2026-02-03 20:48:34 +01:00
parent 9a587dcc4c
commit 36280a6ff8
3 changed files with 186 additions and 122 deletions

View File

@@ -178,8 +178,8 @@ class QualtricsPlotsMixin:
# Use UPPERCASE for category name to distinguish from values
parts.append(f"{display_name.upper()}: {val_str}")
# Get sample size (stored by _ensure_dataframe)
sample_size = getattr(self, '_last_sample_size', None)
# Get sample size from the filtered dataset (not from transformed plot data)
sample_size = self._get_filtered_sample_size()
sample_prefix = f"Sample size: {sample_size}" if sample_size is not None else ""
if not parts:
@@ -297,10 +297,7 @@ class QualtricsPlotsMixin:
return chart
def _ensure_dataframe(self, data: pl.LazyFrame | pl.DataFrame | None) -> pl.DataFrame:
"""Ensure data is an eager DataFrame, collecting if necessary.
Also stores the sample size on self._last_sample_size for use in filter descriptions.
"""
"""Ensure data is an eager DataFrame, collecting if necessary."""
df = data if data is not None else getattr(self, 'data_filtered', None)
if df is None:
raise ValueError("No data provided and self.data_filtered is None.")
@@ -308,9 +305,21 @@ class QualtricsPlotsMixin:
if isinstance(df, pl.LazyFrame):
df = df.collect()
# Store sample size for filter description
self._last_sample_size = df.height
return df
def _get_filtered_sample_size(self) -> int | None:
"""Get the sample size from the filtered dataset (self.data_filtered).
This returns the number of respondents in the filtered dataset,
not the size of any transformed/aggregated data passed to plot functions.
"""
data_filtered = getattr(self, 'data_filtered', None)
if data_filtered is None:
return None
if isinstance(data_filtered, pl.LazyFrame):
return data_filtered.select(pl.len()).collect().item()
return data_filtered.height
def _clean_voice_label(self, col_name: str) -> str:
"""Extract and clean voice name from column name for display.
@@ -681,7 +690,7 @@ class QualtricsPlotsMixin:
ColorPalette.GENDER_FEMALE, ColorPalette.GENDER_FEMALE_NEUTRAL
]
chart = alt.Chart(stats_df).mark_bar().encode(
bars = alt.Chart(stats_df).mark_bar().encode(
x=alt.X('item:N', title=x_label, sort='-y'),
y=alt.Y('count:Q', title=y_label),
color=alt.Color('gender_category:N',
@@ -692,14 +701,27 @@ class QualtricsPlotsMixin:
alt.Tooltip('count:Q', title='1st Place Votes'),
alt.Tooltip('gender:N', title='Gender')
]
).properties(
)
# Text overlay for counts
text = alt.Chart(stats_df).mark_text(
dy=-5,
color='black',
fontSize=10
).encode(
x=alt.X('item:N', sort='-y'),
y=alt.Y('count:Q'),
text=alt.Text('count:Q')
)
chart = (bars + text).properties(
title=self._process_title(title),
width=width or 800,
height=height or getattr(self, 'plot_height', 400)
)
else:
# Bar chart with conditional color
chart = alt.Chart(stats_df).mark_bar().encode(
bars = alt.Chart(stats_df).mark_bar().encode(
x=alt.X('item:N', title=x_label, sort='-y'),
y=alt.Y('count:Q', title=y_label),
color=alt.Color('category:N',
@@ -710,7 +732,20 @@ class QualtricsPlotsMixin:
alt.Tooltip('item:N', title='Item'),
alt.Tooltip('count:Q', title='1st Place Votes')
]
).properties(
)
# Text overlay for counts
text = alt.Chart(stats_df).mark_text(
dy=-5,
color='black',
fontSize=10
).encode(
x=alt.X('item:N', sort='-y'),
y=alt.Y('count:Q'),
text=alt.Text('count:Q')
)
chart = (bars + text).properties(
title=self._process_title(title),
width=width or 800,
height=height or getattr(self, 'plot_height', 400)
@@ -769,7 +804,7 @@ class QualtricsPlotsMixin:
# Text overlay
text = bars.mark_text(
dy=-5,
color='white',
color='black',
fontSize=11
).encode(
text='Weighted Score:Q'