![2 sample t test minitab 2 sample t test minitab](https://i.ytimg.com/vi/LDF0l7fCqMk/maxresdefault.jpg)
That means, we accept ( i.e fail to reject) the alternate hypothesis (Ha) that there is a significant difference between water samples of Supplier-1 and Supplier-2. Referring to the 2-Sample t-test analysis results shown above, in Figure 3, and since the P-Value = 0.000 < 0.05, we reject the Null Hypothesis (Ho). In total, 15 random samples (each with 3 trials) for each Supplier-1 & Supplier- 2 were collected and the data of two independent water suppliers (as shown in Table -1) were analyzed using Minitab® to see if there was significant difference between the water qualities.įigure-4 Output from Minitab® t-distribution Inference:
![2 sample t test minitab 2 sample t test minitab](https://i.ytimg.com/vi/jf9skFLlsBk/hqdefault.jpg)
Average of 3 trials were recorded as one measured value of TDS for one sample.The digital TDS meter used is shown in Figure 1. Each resident measured the value of Total Dissolved Solids (TDS) 3 times (3 trials) using digital TDS meters. The Measurement System : Two (2) residents from the housing colony were selected randomly and each of them was asked to check the water quality from Supplier-1 and Supplier-2. Ha : There is difference between the water samples of Supplier-1 & Supplier-2 Ho : There is no difference between the water samples of Supplier-1 & Supplier-2 If this is OK with us, then the significance level will be 5% (Alpha,α = 0.05) and the Confidence Level will be 95%. This means, we are willing to take a chance of 5% error in our judgment. Then, let us say we want to be 95% confident in declaring our results. To analyze it, first we need to construct the Null Hypothesis (Ho) and the Alternate Hypothesis (Ha) for the claim. Water samples from both the supplier’s were collected and two residents were picked randomly and were asked to check the quality of water using TDS meters for 15 samples collected from each Supplier. The Analytics Premise : It was recommended to do a 2-Sample t-test on the samples of water from both suppliers to see if there was any significant difference. How would Supplier-1 justify that his water is of better quality than the new supplier in order to retain the contract to supply water regularly to such a huge housing complex?
![2 sample t test minitab 2 sample t test minitab](https://i.ytimg.com/vi/53lRTzuhHZI/hqdefault.jpg)
The Secretary of the Housing Complex Association identified another water supplier (Supplier-2-Comp) who supplies, they claimed, better quality of water than the previous supplier (KSM). A dispute was raised by a few residents saying that the water supplied by KSM was of very poor quality and is not usable for drinking/utility purposes (washing, cleaning, bathroom use etc). The Scenario : KSM Water Suppliers, a drinking and utility (raw) water supplies company, supplies water to a large housing complex (with about 400 families) that purchases water from external sources for domestic drinking /utility purposes through its own fleet of water-tankers. The dispute was resolved using Business Analytics tools to analyze the sample data and arrive at a more informed quantitative decision that would be acceptable to all the stakeholders of the dispute the association office-bearers, the consumers,and the water suppliers. The dispute was that both the suppliers claimed their water quality was better than the other. Business Case : I was referred recently, by a large housing complex resident- owner, to resolve a dispute between two drinking water suppliers.