We generate fully random data, i.e. no interrelatedness whatsoever, in an exponential distribution Y with rate X. Data Y are generated for 200 different rates X between 0 and 3, and the generated scores must be round integers between 0 and 3. For each of the simulated distributions, we estimate network strength under the assumption of a common variable model (average full correlations) and under the assumption of a network model (average of absolute partial correlations). Under the hypothesis that network strength and symptom scores are independent (as is the case in normally distributed data), the varying rate would not result in different network strengths.