Perform MRSCAL in Hamlet II for multidimensional scaling

By Avishek Majumder & Priya Chetty on October 25, 2018

The previous article explained how to perform non-metric multidimensional scaling (MDS) method using MINISSA. This article explains how to perform metric multidimensional scaling method MRSCAL in Hamlet II that stands for metric scaling.

Procedure to perform MRSCAL in Hamlet II

Like MINISSA, MRSCAL method can be applied in order to derive similar results from analogous non-linear weighting procedure. This method applies a logarithmic transformation to the distances obtained from the matrix of similarities.

It also follows the same steps as MINISSA up to step 2. Thereafter, Hamlet II will present a log transformation upon selecting the metric ‘MRSCAL method. Click on ‘Scale these items’.

Figure 1: Scales for MRSCAL in Hamlet II
Figure 1: Scales for MRSCAL in Hamlet II

A graph or plot will emerge.

Figure 2: Results of MRSCAL in Hamlet II
Figure 2: Results of MRSCAL in Hamlet II

Editing the graph

This data file offers the coefficients of monotonicity and alienation. It also shows the iterations run to derive the results. This data file also shows the configurations of the graph plotted on the main output window. Moreover, it allows the user to view and analyze all three clusters used while plotting the graph.

In order to edit the data in the graph, click on ‘Data’  as shown in the figure above. In order to draw and toggle the graph, choose the options in the output window as shown in the figure above.

Click on the “Save Display” option to save the edited plot. Hamlet offers to save the MDS results upon exiting the main output window. Save the results if required for further analysis.

INDSCAL for comparing co-occurrence matrices

This article dealt with the application of the metric multi-dimensional scaling method of MRSCAL for text analysis. It helps in visualizing the relationship taking the matrix standardized joint frequencies as its input. Its’s only advantage over MINISSA is that the graphical presentation is not based on corresponding labelled points. It allows partitioned matrices to be presented in 3-D. However, there are other multi-dimensional methods like INDSCAL (Individual Differences Scaling) to compare the acquired co-occurrence matrices. In the next article, application of INDSCAL will be shown and explored in depth.

Priya is the co-founder and Managing Partner of Project Guru, a research and analytics firm based in Gurgaon. She is responsible for the human resource planning and operations functions. Her expertise in analytics has been used in a number of service-based industries like education and financial services.

Her foundational educational is from St. Xaviers High School (Mumbai). She also holds MBA degree in Marketing and Finance from the Indian Institute of Planning and Management, Delhi (2008).

Some of the notable projects she has worked on include:

  • Using systems thinking to improve sustainability in operations: A study carried out in Malaysia in partnership with Universiti Kuala Lumpur.
  • Assessing customer satisfaction with in-house doctors of Jiva Ayurveda (a project executed for the company)
  • Predicting the potential impact of green hydrogen microgirds (A project executed for the Government of South Africa)

She is a key contributor to the in-house research platform Knowledge Tank.

She currently holds over 300 citations from her contributions to the platform.

She has also been a guest speaker at various institutes such as JIMS (Delhi), BPIT (Delhi), and SVU (Tirupati).