Chart Types

Review this article to learn more about all the types of charts supported in Crosschq Insights.

In the Overview of Crosschq Insights article, you discovered the various features available for each chart, including drill-downs, filtering options, and the ability to apply trending data.

This article will introduce you to the different types of charts offered by Crosschq, enhancing your understanding of what each chart represents as you navigate through Insights.

You can always explore the Chart library in Insights to see the full range of available charts.

Click on the hyperlink to navigate directly to the detailed information for each specific chart type:

• Key Performance Indicators (KPI) or Metric Cards
• Vertical Bar Chart
• Horizontal Bar Chart
• Stacked Vertical and Horizontal Bar Charts
• HeatMap Chart
• Donut Chart
• Line Chart
• Tabular Data Summary (Table)
• Scatter Plot with Regression Line
• Coefficient Plot

Key Performance Indicators (KPI) or Metric Cards

Description: Metric cards display key numerical insights or performance indicators in a clean and concise format. They can show absolute values, percentages, or averages, making highlighting important metrics at a glance easy.

Vertical Bar Chart

Description: This chart type uses vertical bars to compare values across categories or time periods. It's great for identifying trends or variations in data over time or between groups.

Horizontal Bar Chart

Description: Horizontal bar charts organize data in descending or ascending order, making it easy to compare categories. This format is particularly useful when category names are long or when emphasis is on ranking.

Stacked Vertical and Horizontal Bar Charts

Description: A stacked bar chart shows the composition of categories within each bar, allowing for a breakdown of data while maintaining a focus on total values. It helps illustrate proportions within groups over time or across categories.

HeatMap Chart

Description: The donut chart visualizes data proportions as segments of a circular shape, similar to a pie chart, but with a center cut out for a cleaner, more modern look. It's ideal for showing percentage breakdowns of a whole.

Line Chart

Description: Line charts are ideal for visualizing trends over time or across a sequential range. Multiple lines can be used to compare performance across groups or categories.

Tabular Data Summary (Table)

Description: This format provides a structured table to present key performance metrics or aggregated data. It's ideal for summarizing details in a way that allows for quick comparisons.

Scatter Plot with Regression Line

Description: Scatter plots use points to represent data for two variables, showing the relationship between them. Each dot represents an individual data point. The regression line overlaid on the plot helps visualize the overall trend or correlation, with statistical details like R² and p-values indicating the strength and significance of the relationship. This chart type is particularly useful for examining relationships between variables, identifying trends or correlations, and determining the presence of a linear relationship between them.

Example from the image:

The scatter plot illustrates the relationship between Age at Hire (x-axis) and the Quality of Hire (QoH) value for Baseline (tenure only) (y-axis). Here's what the data reveals:

• Individual Data Points:
  Each dot represents an individual hire, plotting their age at the time of hire against their QoH value. This helps visualize how QoH varies with age.

• Trend Line (Regression Line):
  The blue line represents the regression trend, summarizing the relationship between age at hire and QoH value. In this case, the trend appears mostly flat, indicating little to no strong correlation between the two variables.

• Statistical Details:

  • R² (Coefficient of Determination): An R² value of 0.003 suggests an extremely weak relationship between age and QoH. This means that age explains only 0.3% of the variance in QoH.
  • P-Value: A p-value of 0.378 indicates that the relationship is not statistically significant, meaning the observed trend is likely due to random chance.

• Outliers:
  Some points are positioned far from the overall trend line, representing individuals whose QoH significantly deviates from the expected pattern. These could indicate exceptional hires (high QoH) or underperformers (low QoH) within specific age groups. It is important to verify the accuracy of these points, as outliers can significantly influence the results of the linear regression.

In summary, the data suggests that Age at Hire has minimal to no impact on QoH based on this dataset. However, outliers may still warrant further investigation for unique insights.

Coefficient Plot

Description: A coefficient plot visually represents the impact of different variables on an outcome. Each dot indicates how much the outcome changes for a specific category compared to a reference group, while horizontal lines represent the confidence intervals. Dots to the right of zero suggest a positive effect, and those to the left indicate a negative effect. If the confidence interval crosses zero, it suggests the effect is not statistically significant, indicating no meaningful difference from the reference group.

Example from the image:

The chart displays the impact of different interviewers on a quality-of-hire (QoH) score, measured relative to a baseline reference group (Abbie Nikolaus, with a mean QoH score of 91.83).

1. Interviewer Names:
   Each row corresponds to an individual interviewer whose influence on the QoH score is being analyzed.

2. Coefficient Values:
   The black dots represent the estimated impact of each interviewer. Positive values (to the right of zero) indicate a positive correlation with QoH, while negative values (to the left of zero) indicate a negative correlation.

3. Confidence Intervals:
   The blue bars show the confidence intervals around each coefficient value. These indicate the range within which the true impact is likely to fall:

   • If the interval crosses zero, the effect is not statistically significant.
   • If the interval stays entirely positive or negative, the effect is statistically significant.

4. Baseline Comparison:
   The baseline is set to the reference group's average QoH score. All impacts are expressed as differences relative to this baseline.

Takeaways:

• Amani Watsica has the highest positive impact (+8.17) with a significant confidence interval, indicating strong positive influence.
• Aiyana Runolfsdottir has a negative impact (-2.43), with statistical significance.
• Antonietta Schuppe has a near-zero impact (+0.48) with a confidence interval crossing zero, suggesting no significant effect.

This data helps identify which interviewers contribute positively or negatively to the quality of hire, guiding performance evaluations and training opportunities.