Robustness checks

In this section I conduct further anal3 es to test and refute alternative explanations of my findings. Two specific explanations come to mind  

To test these two alternative explanations, I conduct two-step regression analj es 1 use lead userness and control variables in a baseline model that is then augmented by functional affiliation and job satisfoction, respectively. Note that the sample size is smaller in the case of job satisfaction [n=97], the reason being that questions on job satisfaction were optional for firms 2, 3 and 4, and were not included at all in the questionnaire used in firm 1. As we will see, my results do not lend empirical support to either of the two alternative explanations. 

by way of a second robustness check, 1 test whether job satisfaction rather than lead userness explains the effects that I discovered of lead userness on organizational behavior. 

1 used simple correlation analysis to explore the relationship between lead userness of employees and job satisfaction. 1 did not find a significant relationship (r=-0,026 p 0,05). This finding came somewhat to my surprise. After all, it would seem natural to suppose that lead users who manage to make their passion their profession should experience higher levels of job satisfaction. 

To the best of my knowdedge, this study is the first to research embedded lead user s innovation-related behavior. 1 find that embedding lead users in producer firms can be an effective way for firms to absorb and leverage user knowdedge for iimovation embedded lead users join need knowdedge, derived from their personal use experience or their interactions with other users, and solution knowledge available inside the different departments of the firm. Thus, 1 suggest th play a crucial role in the iimovation process. 

---

Bock C, Gowik P, Stachel C, Matrix-comprehensive in-house validation and robustness check of a confirmatory method for the determination of four nitrofuran metabolites in poultry muscle and shrimp by LC-MS/MS, J. Chromatogr. B 2007 856 178-189. 

Kruskal-Wallis tests were conducted to find out, if the context (firm affiliation, department affiliation) influenced the responses. This test was appropriate due to the non-normality on item level of most of my data. There were no significant firm or department level differences with respect to the data (5% level of significance), except for five cases (firm affiliation had a significant effect on UEl, Cll, and CI2 department affiliation had a significant effect on CEl and CE2). However this effect seems negligible since the other 49 indicators were not Influenced by the context variables. Yet 1 will control for this effect and its influence on the dependent variables in a robustness check alter the main data analysis. 

Integrity (or robustness) checking ensures that the software will execute predictably. In the event of an integrity failure all software running on a processor... 

Heater Robustness. Check with the reflow oven manufactnrer to ensnre that the heater and fan assemblies (heater modules) will be compatible with the rigors of the higher oven temperatnres associated with Pb-free reflow. Although an oven may be able to reach the higher temperatnres, the heater modules may degrade or fail as a result of running close to their maximnm rated ontpnt. 

Measuring electrodes for impressed current protection are robust reference electrodes (see Section 3.2 and Table 3-1) which are permanently exposed to seawater and remain unpolarized when a small control current is taken. The otherwise usual silver-silver chloride and calomel reference electrodes are used only for checking (see Section 16.7). All reference electrodes with electrolytes and diaphragms are unsuitable as long-term electrodes for potential-controlled rectifiers. Only metal-medium electrodes which have a sufficiently constant potential can be considered as measuring electrodes. The silver-silver chloride electrode has a potential that depends on the chloride content of the water [see Eq. (2-29)]. This potential deviation can usually be tolerated [3]. The most reliable electrodes are those of pure zinc [3]. They have a constant rest potential, are slightly polarizable and in case of film formation can be regenerated by an anodic current pulse. They last at least 5 years. 

It does not calculate source emission rates. While it handles jets, it does so simply and docs not calculate the details of the jet motions and thermodynamics. It should not be used for strongly buoyant plumes. The error diagnostics are limited to checking the consistency of input parameters. Run time error diagnostics are missing but are rarely needed due to its robustness. 

Table 2 The steps of recognizing a problem, proposing a solution, checking it for robust operation, and documenting the procedure and results under GMP are nested operations...

The most useful characteristic of the median is the small influence exerted on it by extreme values, that is, its robust nature. The median can thus serve as a check on the calculated mean. 

OECD also provides a check list for the application of its principles in the context of QSAR validation. This checklist can be useful to help scientists and regulators during the selection of a QSAR model and to evaluate its robustness/ validity [9]. 

Multifactorial experiments are used in analytical chemistry for diverse applications, e.g., checking up significant influences before optimization procedures, recognizing matrix effects, and testing the robustness of analytical procedures (Wegscheider [1996]). 

In such cases it may be possible to check on robustness and ruggedness by means of statistical tests (see Sect. 4.3). All the variations to the measured signal, apart from that of the analyte, can be considered in form of error terms see Eq. (3.6) ... 

If the validated test method requires 1 g of material but only 100 mg is available, you must find out if the method is sufficiently robust to stand this amount of scaling down. This has to be checked before the analysis starts, i.e. the method must be validated for analysis of 100 mg of material. Even if the method of analysis is found to be robust, scaling down is only a viable option if the smaller test portion size remains representative, within acceptable limits. This will depend on the homogeneity of the material. 

Additives are often used to increase selectivity. They are paramount in chiral separations, but they are also frequently used in non-chiral separations, e.g., cyclodextrins (CDs). In our lab, BGEs with and without a cyclodextrin are part of our generic protocol. Figure 8 demonstrates that although one can more or less predict interaction with the additive from the chemical structures, it is still difficult to predict separation.Batch-to-batch variability and variability between suppliers can be a problem of (chiral) additives and a check of different batches has to be part of the robustness test (e.g., reference 56). If the additive is charged and has one or more pK s around the pH of the BGE, extra care should be taken to control the pH. Alternatively, better robustness might be obtained with another uncharged additive, even if this results in lower resolution. 

A check of robustness includes preliminary experiments on precision. During robustness testing, a single standard is repetitively analyzed before starting the actual calibration. Without sufficient precision at a single concentration, it is fruitless to calibrate. However, the RSD obtained from repeated injections underestimates the overall error by a factor of up to 3. ... 

It has already been mentioned that control of a continuous process includes control and monitoring of raw materials used in the process. This is a very successful area of application for NIR because of the minimal sample preparation required, the robustness of the instrumentation and its ability to measure both chemical and physical properties [117]. Libraries of raw material can be compiled, which can then be shared between manufacturing locations given adequate calibration checks between measuring equipment [118]. 

The instruments for polymer HPLC except for the columns (Section 16.8.1) and for some detectors are in principle the same as for the HPLC of small molecules. Due to sensitivity of particular detectors to the pressure variations (Section 16.9.1) the pumping systems should be equipped with the efficient dampeners to suppress the rest pulsation of pressure and flow rate of mobile phase. In most methods of polymer HPLC, and especially in SEC, the retention volume of sample (fraction) is the parameter of the same importance as the sample concentration. The conventional volumeters— siphons, drop counters, heat pulse counters—do not exhibit necessary robustness and precision [270]. Therefore the timescale is utilized and the eluent flow rate has to be very constant even when rather viscous samples are introduced into column. The problems with the constant eluent flow rate may be caused by the poor resettability of some pumping systems. Therefore, it is advisable to carefully check the actual flow rate after each restarting of instrument and in the course of the long-time experiments. A continuous operation— 24h a day and 7 days a week—is advisable for the high-precision SEC measurements. THE or other eluent is continuously distilled and recycled. 

In pharmaceutical technology, quality assurance of the pharmaceutical formulation is important. When a pharmaceutical formulation is produced, on-line quality monitoring and control has to be performed in order to check the quality of the outgoing products. Methodology to perform this task is Statistical Process Control (SPC) and is not included in this book. Good text books in the area of SPC exists [6-9]. In this book the focus is on off-line quality control, e.g. how to make products that are intrinsic robust against process variations. 

If robustness has to be build in, then the concept of robustness has to be formalized and optimized. This is contrary to the class of methods that check the robustness or ruggedness of existing methods then the influence... 

---